Dmitry Ryumin

DmitryRyumin

AI & ML interests

Machine Learning and Applications, Multi-Modal Understanding

Organizations

DmitryRyumin's activity

posted an update 8 days ago
view post
Post
2074
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing ๐Ÿ”

๐Ÿ“ Description: IntrinsicAvatar is a method for extracting high-quality geometry, albedo, material, and lighting properties of clothed human avatars from monocular videos using explicit ray tracing and volumetric scattering, enabling realistic animations under varying lighting conditions.

๐Ÿ‘ฅ Authors: Shaofei Wang, Boลพidar Antiฤ‡, Andreas Geiger, and Siyu Tang

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing (2312.05210)

๐ŸŒ Github Page: https://neuralbodies.github.io/IntrinsicAvatar/
๐Ÿ“ Repository: https://github.com/taconite/IntrinsicAvatar

๐Ÿ“บ Video: https://www.youtube.com/watch?v=aS8AIxgVXzI

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿš€ WACV-2024-Papers: https://github.com/DmitryRyumin/WACV-2024-Papers

๐Ÿš€ ICCV-2023-Papers: https://github.com/DmitryRyumin/ICCV-2023-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #IntrinsicAvatar #InverseRendering #MonocularVideos #RayTracing #VolumetricScattering #3DReconstruction #MachineLearning #ComputerVision #DeepLearning #AI #CVPR2024
posted an update 9 days ago
view post
Post
2912
๐Ÿ”ฅ๐ŸŽญ๐ŸŒŸ New Research Alert - ECCV 2024 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿ”ฅ
๐Ÿ“„ Title: RodinHD: High-Fidelity 3D Avatar Generation with Diffusion Models ๐Ÿ”

๐Ÿ“ Description: RodinHD generates high-fidelity 3D avatars from portrait images using a novel data scheduling strategy and weight consolidation regularization to capture intricate details such as hairstyles.

๐Ÿ‘ฅ Authors: Bowen Zhang, @yiji , @chunyuwang , Ting Zhang, @jiaolong , Yansong Tang, Feng Zhao, Dong Chen, and Baining Guo

๐Ÿ“… Conference: ECCV, 29 Sep โ€“ 4 Oct, 2024 | Milano, Italy ๐Ÿ‡ฎ๐Ÿ‡น

๐Ÿ“„ Paper: RodinHD: High-Fidelity 3D Avatar Generation with Diffusion Models (2407.06938)

๐ŸŒ Github Page: https://rodinhd.github.io/
๐Ÿ“ Repository: https://github.com/RodinHD/RodinHD

๐Ÿ“บ Video: https://www.youtube.com/watch?v=ULvHt7dZx-Q

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿš€ WACV-2024-Papers: https://github.com/DmitryRyumin/WACV-2024-Papers

๐Ÿš€ ICCV-2023-Papers: https://github.com/DmitryRyumin/ICCV-2023-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #RodinHD #3DAvatars #DiffusionModels #HighFidelity #PortraitTo3D #MachineLearning #ComputerVision #DeepLearning #AI #ECCV2024
posted an update 13 days ago
view post
Post
2357
๐Ÿ”ฅ๐ŸŽญ๐ŸŒŸ New Research Alert - LivePortrait (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿ”ฅ
๐Ÿ“„ Title: LivePortrait: Efficient Portrait Animation with Stitching and Retargeting Control ๐Ÿ”

๐Ÿ“ Description: LivePortrait is an efficient video-driven portrait animation framework that uses implicit keypoints and stitching/retargeting modules to generate high-quality, controllable animations from a single source image.

๐Ÿ‘ฅ Authors: @cleardusk , Dingyun Zhang, Xiaoqiang Liu, Zhizhou Zhong, Yuan Zhang, Pengfei Wan, and Di Zhang

๐Ÿค— Demo: KwaiVGI/LivePortrait

๐Ÿ“„ Paper: LivePortrait: Efficient Portrait Animation with Stitching and Retargeting Control (2407.03168)

๐ŸŒ Github Page: https://liveportrait.github.io/
๐Ÿ“ Repository: https://github.com/KwaiVGI/LivePortrait

๐Ÿ”ฅ Model ๐Ÿค–: KwaiVGI/LivePortrait

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿš€ WACV-2024-Papers: https://github.com/DmitryRyumin/WACV-2024-Papers

๐Ÿš€ ICCV-2023-Papers: https://github.com/DmitryRyumin/ICCV-2023-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #LivePortrait #PortraitAnimation #ComputerVision #MachineLearning #DeepLearning #ComputerGraphics #FacialAnimation #GenerativeAI #RealTimeRendering #AI
posted an update 18 days ago
view post
Post
2691
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert (Avatars Collection)! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: Expressive Gaussian Human Avatars from Monocular RGB Video ๐Ÿ”

๐Ÿ“ Description: The new EVA model enhances the expressiveness of digital avatars by using 3D Gaussians and SMPL-X to capture fine-grained hand and face details from monocular RGB video.

๐Ÿ‘ฅ Authors: Hezhen Hu, Zhiwen Fan, Tianhao Wu, Yihan Xi, Seoyoung Lee, Georgios Pavlakos, and Zhangyang Wang

๐Ÿ“„ Paper: Expressive Gaussian Human Avatars from Monocular RGB Video (2407.03204)

๐ŸŒ Github Page: https://evahuman.github.io/
๐Ÿ“ Repository: https://github.com/evahuman/EVA

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿš€ WACV-2024-Papers: https://github.com/DmitryRyumin/WACV-2024-Papers

๐Ÿš€ ICCV-2023-Papers: https://github.com/DmitryRyumin/ICCV-2023-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #DigitalAvatars #3DModeling #ComputerVision #MonocularVideo #SMPLX #3DGaussians #AvatarExpressiveness #HandTracking #FacialExpressions #AI #MachineLearning
posted an update 21 days ago
view post
Post
2016
๐Ÿ”ฅ๐ŸŽญ๐ŸŒŸ New Research Alert - ECCV 2024 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿ”ฅ
๐Ÿ“„ Title: Topo4D: Topology-Preserving Gaussian Splatting for High-Fidelity 4D Head Capture ๐Ÿ”

๐Ÿ“ Description: Topo4D is a novel method for automated, high-fidelity 4D head tracking that optimizes dynamic topological meshes and 8K texture maps from multi-view time-series images.

๐Ÿ‘ฅ Authors: @Dazz1e , Y. Cheng, @Ryan-sjtu , H. Jia, D. Xu, W. Zhu, Y. Yan

๐Ÿ“… Conference: ECCV, 29 Sep โ€“ 4 Oct, 2024 | Milano, Italy ๐Ÿ‡ฎ๐Ÿ‡น

๐Ÿ“„ Paper: Topo4D: Topology-Preserving Gaussian Splatting for High-Fidelity 4D Head Capture (2406.00440)

๐ŸŒ Github Page: https://xuanchenli.github.io/Topo4D/
๐Ÿ“ Repository: https://github.com/XuanchenLi/Topo4D

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿš€ WACV-2024-Papers: https://github.com/DmitryRyumin/WACV-2024-Papers

๐Ÿš€ ICCV-2023-Papers: https://github.com/DmitryRyumin/ICCV-2023-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #Topo4D #4DHead #3DModeling #4DCapture #FacialAnimation #ComputerGraphics #MachineLearning #HighFidelity #TextureMapping #DynamicMeshes #GaussianSplatting #VisualEffects #ECCV2024
  • 1 reply
ยท
posted an update 26 days ago
view post
Post
3539
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - Portrait4D-v2 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: Portrait4D-v2: Pseudo Multi-View Data Creates Better 4D Head Synthesizer ๐Ÿ”

๐Ÿ“ Description: Portrait4D-v2 is a novel method for one-shot 4D head avatar synthesis using pseudo multi-view videos and a vision transformer backbone, achieving superior performance without relying on 3DMM reconstruction.

๐Ÿ‘ฅ Authors: Yu Deng, Duomin Wang, and Baoyuan Wang

๐Ÿ“„ Paper: Portrait4D-v2: Pseudo Multi-View Data Creates Better 4D Head Synthesizer (2403.13570)

๐ŸŒ GitHub Page: https://yudeng.github.io/Portrait4D-v2/
๐Ÿ“ Repository: https://github.com/YuDeng/Portrait-4D

๐Ÿ“บ Video: https://www.youtube.com/watch?v=5YJY6-wcOJo

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: Portrait4D #4DAvatar #HeadSynthesis #3DModeling #TechInnovation #DeepLearning #ComputerGraphics #ComputerVision #Innovation
  • 1 reply
ยท
posted an update about 1 month ago
view post
Post
2338
๐Ÿ˜€๐Ÿ˜ฒ๐Ÿ˜๐Ÿ˜ก New Research Alert - CVPRW 2024 (Facial Expressions Recognition Collection)! ๐Ÿ˜ก๐Ÿ˜ฅ๐Ÿฅด๐Ÿ˜ฑ
๐Ÿ“„ Title: Zero-Shot Audio-Visual Compound Expression Recognition Method based on Emotion Probability Fusion ๐Ÿ”

๐Ÿ“ Description: AVCER is a novel audio-visual method for compound expression recognition based on pair-wise sum of emotion probability, evaluated in multi- and cross-corpus setups without task-specific training data, demonstrating its potential for intelligent emotion annotation tools.

๐Ÿ‘ฅ Authors: @ElenaRyumina , Maxim Markitantov, @DmitryRyumin , Heysem Kaya, and Alexey Karpov

๐Ÿ“… Conference: CVPRW, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿค— Demo: ElenaRyumina/AVCER

๐Ÿ“„ Paper: Audio-Visual Compound Expression Recognition Method based on Late Modality Fusion and Rule-based Decision (2403.12687)

๐ŸŒ Github Page: https://elenaryumina.github.io/AVCER
๐Ÿ“ Repository: https://github.com/ElenaRyumina/AVCER/tree/main/src

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Facial Expressions Recognition Collection: DmitryRyumin/facial-expressions-recognition-65f22574e0724601636ddaf7

๐Ÿ” Keywords: #AVCER #AudioVisual #CompoundExpressions #EmotionRecognition #ModalityFusion #RuleBasedAI #ABAWCompetition #AIResearch #HumanEmotion #IntelligentTools #MachineLearning #DeepLearning #MultiCorpus #CrossCorpus #CVPR2024
posted an update about 2 months ago
view post
Post
1853
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: Relightable Gaussian Codec Avatars ๐Ÿ”

๐Ÿ“ Description: Relightable Gaussian Codec Avatars is a method for creating highly detailed and relightable 3D head avatars that can animate expressions in real time and support complex features such as hair and skin with efficient rendering suitable for VR.

๐Ÿ‘ฅ Authors: @psyth , @GBielXONE02 , Tomas Simon, Junxuan Li, and @giljoonam

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: Relightable Gaussian Codec Avatars (2312.03704)

๐ŸŒ GitHub Page: https://shunsukesaito.github.io/rgca/

๐Ÿš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #3DAvatars #RealTimeRendering #RelightableAvatars #3DModeling #VirtualReality #CVPR2024 #DeepLearning #ComputerGraphics #ComputerVision #Innovation #VR
posted an update about 2 months ago
view post
Post
802
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - InstructAvatar (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation ๐Ÿ”

๐Ÿ“ Description: InstructAvatar is a novel method for generating emotionally expressive 2D avatars using text-guided instructions, offering improved emotion control, lip-sync quality, and naturalness. It uses a two-branch diffusion-based generator to predict avatars based on both audio and text input.

๐Ÿ‘ฅ Authors: Yuchi Wang et al.

๐Ÿ“„ Paper: InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation (2405.15758)

๐ŸŒ Github Page: https://wangyuchi369.github.io/InstructAvatar/
๐Ÿ“ Repository: https://github.com/wangyuchi369/InstructAvatar

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #InstructAvatar #AvatarGeneration #EmotionControl #FacialMotion #LipSynchronization #NaturalLanguageInterface #DiffusionBasedGenerator #TextGuidedInstructions #2DAvatars #VideoSynthesis #Interactivity #ComputerGraphics #DeepLearning #ComputerVision #Innovation
posted an update about 2 months ago
view post
Post
1470
๐Ÿ”ฅ๐Ÿš€๐ŸŒŸ New Research Alert - YOLOv10! ๐ŸŒŸ๐Ÿš€๐Ÿ”ฅ
๐Ÿ“„ Title: YOLOv10: Real-Time End-to-End Object Detection ๐Ÿ”

๐Ÿ“ Description: YOLOv10 improves real-time object recognition by eliminating non-maximum suppression and optimizing the model architecture to achieve state-of-the-art performance with lower latency and computational overhead.

๐Ÿ‘ฅ Authors: Ao Wang et al.

๐Ÿ“„ Paper: YOLOv10: Real-Time End-to-End Object Detection (2405.14458)

๐Ÿค— Demo: kadirnar/Yolov10 curated by @kadirnar
๐Ÿ”ฅ Model ๐Ÿค–: kadirnar/Yolov10

๐Ÿ“ Repository: https://github.com/THU-MIG/yolov10

๐Ÿ“ฎ Post about YOLOv9 - https://huggingface.co/posts/DmitryRyumin/519784698531054

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #YOLOv10 #ObjectDetection #RealTimeAI #ModelOptimization #MachineLearning #DeepLearning #ComputerVision #Innovation
  • 1 reply
ยท
posted an update 2 months ago
view post
Post
1533
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - Gaussian Head & Shoulders (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping ๐Ÿ”

๐Ÿ“ Description: Gaussian Head & Shoulders is a method for creating high-fidelity upper body avatars by integrating 3D morphable head models with a neural texture warping approach to overcome the limitations of Gaussian splatting.

๐Ÿ‘ฅ Authors: Tianhao Wu et al.

๐Ÿ“„ Paper: Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping (2405.12069)

๐ŸŒ Github Page: https://gaussian-head-shoulders.netlify.app

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #3DModeling #NeuralAvatars #GaussianSplatting #HighFidelityAvatars #3DReconstruction #AvatarRendering #TextureWarping #ComputerGraphics #DeepLearning #ComputerVision #Innovation
posted an update 2 months ago
view post
Post
1752
๐Ÿš€๐Ÿค–๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿค–๐Ÿš€
๐Ÿ“„ Title: RoHM: Robust Human Motion Reconstruction via Diffusion ๐Ÿ”

๐Ÿ“ Description: RoHM is a diffusion-based approach for robust 3D human motion reconstruction from monocular RGB(-D) videos, effectively handling noise and occlusions to produce complete and coherent motions. This method outperforms current techniques in various tasks and is faster at test time.

๐Ÿ‘ฅ Authors: Siwei Zhang et al.

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: RoHM: Robust Human Motion Reconstruction via Diffusion (2401.08570)

๐ŸŒ GitHub Page: https://sanweiliti.github.io/ROHM/ROHM.html
๐Ÿ“ Repository: https://github.com/sanweiliti/RoHM

๐Ÿš€ Added to the CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #RoHM #HumanMotionReconstruction #DiffusionModels #3DAnimation #CVPR2024 #DeepLearning #ComputerVision #Innovation
posted an update 2 months ago
view post
Post
1039
๐Ÿš€๐Ÿ‘•๐ŸŒŸ New Research Alert - SIGGRAPH 2024 (Avatars Collection)! ๐ŸŒŸ๐Ÿ‘š๐Ÿš€
๐Ÿ“„ Title: LayGA: Layered Gaussian Avatars for Animatable Clothing Transfer ๐Ÿ”

๐Ÿ“ Description: LayGA is a novel method for animatable clothing transfer that separates the body and clothing into two layers for improved photorealism and accurate clothing tracking, outperforming existing methods.

๐Ÿ‘ฅ Authors: Siyou Lin, Zhe Li, Zhaoqi Su, Zerong Zheng, Hongwen Zhang, and Yebin Liu

๐Ÿ“… Conference: SIGGRAPH, 28 Jul โ€“ 1 Aug, 2024 | Denver CO, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: LayGA: Layered Gaussian Avatars for Animatable Clothing Transfer (2405.07319)

๐ŸŒ Github Page: https://jsnln.github.io/layga/index.html

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #LayGA #AnimatableClothingTransfer #VirtualTryOn #AvatarTechnology #SIGGRAPH2024 #ComputerGraphics #DeepLearning #ComputerVision #Innovation
replied to their post 2 months ago
posted an update 2 months ago
view post
Post
1269
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - AniTalker (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: AniTalker: Animate Vivid and Diverse Talking Faces through Identity-Decoupled Facial Motion Encoding ๐Ÿ”

๐Ÿ“ Description: AniTalker is a new framework that transforms a single static portrait and a single input audio file into animated, talking videos with natural, fluid movements.

๐Ÿ‘ฅ Authors: Tao Liu, Feilong Chen, Shuai Fan, @cpdu , Qi Chen, Xie Chen, and Kai Yu

๐Ÿ“„ Paper: AniTalker: Animate Vivid and Diverse Talking Faces through Identity-Decoupled Facial Motion Encoding (2405.03121)

๐ŸŒ Github Page: https://x-lance.github.io/AniTalker
๐Ÿ“ Repository: https://github.com/X-LANCE/AniTalker

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #AniTalker #FacialAnimation #DynamicAvatars #FaceSynthesis #TalkingFaces #DiffusionModel #ComputerGraphics #DeepLearning #ComputerVision #Innovation
  • 2 replies
ยท
posted an update 2 months ago
view post
Post
1073
๐Ÿ˜€๐Ÿ˜ฒ๐Ÿ˜๐Ÿ˜ก New Research Alert - FER-YOLO-Mamba (Facial Expressions Recognition Collection)! ๐Ÿ˜ก๐Ÿ˜ฅ๐Ÿฅด๐Ÿ˜ฑ
๐Ÿ“„ Title: FER-YOLO-Mamba: Facial Expression Detection and Classification Based on Selective State Space ๐Ÿ”

๐Ÿ“ Description: FER-YOLO-Mamba is a novel facial expression recognition model that combines the strengths of YOLO and Mamba technologies to efficiently recognize and localize facial expressions.

๐Ÿ‘ฅ Authors: Hui Ma, Sen Lei, Turgay Celik, and Heng-Chao Li

๐Ÿ”— Paper: FER-YOLO-Mamba: Facial Expression Detection and Classification Based on Selective State Space (2405.01828)

๐Ÿ“ Repository: https://github.com/SwjtuMa/FER-YOLO-Mamba

๐Ÿš€ Added to the Facial Expressions Recognition Collection: DmitryRyumin/facial-expressions-recognition-65f22574e0724601636ddaf7

๐Ÿ”ฅ๐Ÿ” See also Facial_Expression_Recognition - ElenaRyumina/Facial_Expression_Recognition (App, co-authored by @DmitryRyumin ) ๐Ÿ˜‰

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #FERYOLOMamba #FER #YOLO #Mamba #FacialExpressionRecognition #EmotionRecognition #ComputerVision #DeepLearning #MachineLearning #Innovation
posted an update 2 months ago
view post
Post
1773
๐Ÿ”ฅ๐Ÿš€๐ŸŒŸ New Research Alert - YOCO! ๐ŸŒŸ๐Ÿš€๐Ÿ”ฅ
๐Ÿ“„ Title: You Only Cache Once: Decoder-Decoder Architectures for Language Models ๐Ÿ”

๐Ÿ“ Description: YOCO is a novel decoder-decoder architecture for LLMs that reduces memory requirements, speeds up prefilling, and maintains global attention. It consists of a self-decoder for encoding KV caches and a cross-decoder for reusing these caches via cross-attention.

๐Ÿ‘ฅ Authors: Yutao Sun et al.

๐Ÿ“„ Paper: You Only Cache Once: Decoder-Decoder Architectures for Language Models (2405.05254)

๐Ÿ“ Repository: https://github.com/microsoft/unilm/tree/master/YOCO

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #YOCO #DecoderDecoder #LargeLanguageModels #EfficientArchitecture #GPUMemoryReduction #PrefillingSpeedup #GlobalAttention #DeepLearning #Innovation #AI
ยท
posted an update 3 months ago
view post
Post
2055
๐Ÿ”ฅ๐Ÿš€๐ŸŒŸ New Research Alert - xLSTM! ๐ŸŒŸ๐Ÿš€๐Ÿ”ฅ
๐Ÿ“„ Title: xLSTM: Extended Long Short-Term Memory ๐Ÿ”

๐Ÿ“ Description: xLSTM is a scaled-up LSTM architecture with exponential gating and modified memory structures to mitigate known limitations. xLSTM blocks outperform SOTA transformers and state-space models in performance and scaling.

๐Ÿ‘ฅ Authors: Maximilian Beck et al.

๐Ÿ“„ Paper: xLSTM: Extended Long Short-Term Memory (2405.04517)

๐Ÿ“ Repository: https://github.com/NX-AI/xlstm

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #xLSTM #DeepLearning #Innovation #AI
  • 1 reply
ยท
posted an update 3 months ago
view post
Post
2577
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - SIGGRAPH 2024 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: 3D Gaussian Blendshapes for Head Avatar Animation ๐Ÿ”

๐Ÿ“ Description: 3D Gaussian Blendshapes for Head Avatar Animation is a novel method for modeling and animating photorealistic head avatars from monocular video input.

๐Ÿ‘ฅ Authors: Shengjie Ma, Yanlin Weng, Tianjia Shao, and Kun Zhou

๐Ÿ“… Conference: SIGGRAPH, 28 Jul โ€“ 1 Aug, 2024 | Denver CO, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: 3D Gaussian Blendshapes for Head Avatar Animation (2404.19398)

๐ŸŒ Github Page: https://gapszju.github.io/GaussianBlendshape/

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #3DAnimation #HeadAvatar #GaussianBlendshapes #FacialAnimation #RealTimeRendering #SIGGRAPH2024 #ComputerGraphics #DeepLearning #ComputerVision #Innovation
replied to their post 3 months ago
view reply

The authors plan to release the Dataset itself in June and the code in July, apparently during CVPR.

posted an update 3 months ago
view post
Post
1374
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: EMOPortraits: Emotion-enhanced Multimodal One-shot Head Avatars ๐Ÿ”

๐Ÿ“ Description: EMOPortraits is an enhanced multimodal one-shot head avatar model that achieves SOTA performance in emotion transfer and audio-driven facial animation tasks by improving the training pipeline and architecture to better handle intense and asymmetric facial expressions, while also proposing a novel multiview video dataset containing a wide range of such expressions.

๐Ÿ‘ฅ Authors: Nikita Drobyshev, Antoni Bigata Casademunt, Konstantinos Vougioukas, Zoe Landgraf, Stavros Petridis, and Maja Pantic

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: EMOPortraits: Emotion-enhanced Multimodal One-shot Head Avatars (2404.19110)

๐ŸŒ GitHub Page: https://neeek2303.github.io/EMOPortraits

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #EMOPortraits #EmotionalTransfer #FacialAnimation #HeadAvatar #MultimodalLearning #OneShotLearning #AsymmetricFacialExpressions #IntenseFacialExpressions #NovelDataset #CVPR2024 #DeepLearning #ComputerVision #Innovation
  • 3 replies
ยท
posted an update 3 months ago
view post
Post
1481
๐Ÿš€๐ŸŽญ๐Ÿ”ฅ New Research Alert (Avatars Collection)! ๐Ÿ”ฅ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: ConsistentID: Portrait Generation with Multimodal Fine-Grained Identity Preserving ๐Ÿ”

๐Ÿ“ Description: ConsistentID is a novel portrait generation method that preserves the fine-grained identity of a single reference image.

๐Ÿ‘ฅ Authors: Jiehui Huang et al.

๐Ÿ”— Paper: ConsistentID: Portrait Generation with Multimodal Fine-Grained Identity Preserving (2404.16771)

๐ŸŒ Github Page: https://ssugarwh.github.io/consistentid.github.io/
๐Ÿ“ Repository: https://github.com/JackAILab/ConsistentID

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #ConsistentID #PortraitGeneration #IdentityPreservation #FineGrainedControl #ImageSynthesis #GenerativeModels #ComputerVision #DeepLearning
posted an update 3 months ago
view post
Post
2160
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: WANDR: Intention-guided Human Motion Generation ๐Ÿ”

๐Ÿ“ Description: WANDR is a conditional Variational AutoEncoder (c-VAE) that generates realistic motion of human avatars that navigate towards an arbitrary goal location and reach for it.

๐Ÿ‘ฅ Authors: Markos Diomataris, Nikos Athanasiou, Omid Taheri, Xi Wang, Otmar Hilliges, Michael J. Black

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ“„ Paper: WANDR: Intention-guided Human Motion Generation (2404.15383)

๐ŸŒ Web Page: https://wandr.is.tue.mpg.de
๐Ÿ“ Repository: https://github.com/markos-diomataris/wandr

๐Ÿ“บ Video: https://www.youtube.com/watch?v=9szizM-XUCg

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #WANDR #HumanMotionGeneration #MotionSynthesis #3DAvatar #GoalOrientedMovement #IntentionGuided #ConditionalVAE #CVPR2024 #DeepLearning #Innovation
posted an update 3 months ago
view post
Post
1906
๐Ÿš€๐ŸŽญ๐Ÿ”ฅ New Research Alert (Avatars Collection)! ๐Ÿ”ฅ๐Ÿ‘„๐Ÿš€
๐Ÿ“„ Title: Learn2Talk: 3D Talking Face Learns from 2D Talking Face

๐Ÿ“ Description: Learn2Talk is a framework that leverages expertise from 2D talking face methods to improve 3D talking face synthesis, focusing on lip synchronization and speech perception.

๐Ÿ‘ฅ Authors: Yixiang Zhuang et al.

๐Ÿ”— Paper: Learn2Talk: 3D Talking Face Learns from 2D Talking Face (2404.12888)

๐ŸŒ Github Page: https://lkjkjoiuiu.github.io/Learn2Talk/

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #Learn2Talk #3DTalkingFace #SpeechDrivenFacialAnimation #LipSync #SpeechPerception #ComputerVision #ImageProcessing #DeepLearning
posted an update 3 months ago
view post
Post
1778
๐Ÿ˜€๐Ÿค“๐Ÿ˜Ž New Research Alert - NAACL 2024 (Big Five Personality Traits Collection)! ๐Ÿ˜Ž๐Ÿ˜‰๐Ÿ˜ค
๐Ÿ“„ Title: PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits ๐Ÿ’ฌ

๐Ÿ“ Description: This research examines the ability of LLMs to express personality traits and finds that LLMs can generate content consistent with assigned personality profiles and that humans can recognize certain traits with up to 80% accuracy. However, accuracy drops significantly when annotators are aware that the content was generated by an AI.

๐Ÿ‘ฅ Authors: Hang Jiang et al.

๐Ÿ“… Conference: NAACL, June 16โ€“21, 2024 | Mexico City, Mexico ๐Ÿ‡ฒ๐Ÿ‡ฝ

๐Ÿ”— Paper: PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits (2305.02547)

๐Ÿ“ Repository: https://github.com/hjian42/PersonaLLM

๐Ÿš€ Added to the Big Five Personality Traits Collection: DmitryRyumin/big-five-personality-traits-661fb545292ab3d12a5a4890

๐Ÿ”ฅ๐Ÿ” See also OCEAN-AI - ElenaRyumina/OCEANAI (App, co-authored by @DmitryRyumin ) ๐Ÿ˜‰

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #PersonaLLM #OCEANAI #BigFive #PersonalityTraits #PersonalityAnalysis #Chatbots #LLMs #NAACL2024 #DeepLearning #Innovation
posted an update 3 months ago
view post
Post
3052
๐Ÿš€๐Ÿ’‡โ€โ™‚๏ธ๐Ÿ”ฅ New Research Alert (Avatars Collection)! ๐Ÿ”ฅ๐Ÿ’‡โ€โ™€๏ธ๐Ÿš€
๐Ÿ“„ Title: HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach

๐Ÿ“ Description: HairFastGAN is a fast, encoder-based approach to realistic and robust hair transfer that operates in the FS latent space of StyleGAN and includes enhanced in-painting and improved encoders for better alignment, color transfer, and post-processing.

๐Ÿ‘ฅ Authors: Maxim Nikolaev, Mikhail Kuznetsov, Dmitry Vetrov, and Aibek Alanov

๐Ÿ”— Paper: HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach (2404.01094)

๐Ÿ“ Repository: https://github.com/AIRI-Institute/HairFastGAN

๐Ÿค— Demo: multimodalart/hairfastgan
๐Ÿ”ฅ Model ๐Ÿค–: AIRI-Institute/HairFastGAN

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #HairFastGAN #StyleGAN #VirtualTryOn #HairTransfer #AIHairStyling #GenerativeModels #ComputerVision #ImageProcessing #DeepLearning
posted an update 3 months ago
view post
Post
2299
๐Ÿš€๐Ÿ‘ฉโ€๐ŸŽค๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿ‘ฉโ€๐ŸŽค๐Ÿš€
๐Ÿ“„ Title: Generalizable Face Landmarking Guided by Conditional Face Warping

๐Ÿ“ Description: A new method is proposed to learn a generalizable face landmark that can handle different facial styles, using labeled real faces and unlabeled stylized faces.

๐Ÿ‘ฅ Authors: Jiayi Liang, Haotian Liu, Hongteng Xu, Dixin Luo

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: Generalizable Face Landmarking Guided by Conditional Face Warping (2404.12322)

๐ŸŒ Github Page: https://plustwo0.github.io/project-face-landmarker/
๐Ÿ“ Repository: https://github.com/plustwo0/generalized-face-landmarker

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #FaceLandmarking #DomainAdaptation #FaceWarpping #CVPR2024 #DeepLearning #Innovation
replied to Jaward's post 3 months ago
view reply

Thank you so much! I'm planning a series of posts on Personality Traits and the Big Five. All posts will be related to the collection you mentioned. Also, together with @ElenaRyumina , we are planning to expand the App (OCEANAI) and are preparing a publication. I hope to be able to share in the future after the accepted paper to the A-Level Conference.

replied to Jaward's post 3 months ago
posted an update 3 months ago
view post
Post
2938
๐Ÿ˜€๐Ÿค“๐Ÿ˜Ž New Research Alert - LREC-COLING 2024 (Big Five Personality Traits Collection)! ๐Ÿ˜Ž๐Ÿ˜‰๐Ÿ˜ค
๐Ÿ“„ Title: PSYDIAL: Personality-based Synthetic Dialogue Generation using Large Language Models

๐Ÿ“ Description: The PSYDIAL presents a novel pipeline for generating personality-based synthetic dialog data to elicit more human-like responses from language models, and presents a Korean dialog dataset focused on personality-based dialog.

๐Ÿ‘ฅ Authors: Ji-Eun Han et al.

๐Ÿ“… Conference: LREC-COLING, May 20-25, 2024 | Torino, Italia ๐Ÿ‡ฎ๐Ÿ‡น

๐Ÿ”— Paper: PSYDIAL: Personality-based Synthetic Dialogue Generation using Large Language Models (2404.00930)

๐Ÿš€ Added to the Big Five Personality Traits Collection: DmitryRyumin/big-five-personality-traits-661fb545292ab3d12a5a4890

๐Ÿ”ฅ๐Ÿ” See also OCEAN-AI - ElenaRyumina/OCEANAI (App, co-authored by @DmitryRyumin ) ๐Ÿ˜‰

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #PSYDIAL #PersonalityDialogues #SyntheticData #LanguageModels #ConversationalAI #KoreanDialogues #BigFivePersonality #ExtraveresionDialogues #OCEANAI #BigFive #PersonalityTraits #PersonalityAnalysis #LREC-COLING2024 #DeepLearning #Innovation
posted an update 3 months ago
view post
Post
2450
๐Ÿ˜€๐Ÿค“๐Ÿ˜Ž New Space - OCEAN-AI (App, co-authored by @DmitryRyumin ) ๐Ÿ˜Ž๐Ÿ˜‰๐Ÿ˜ค
๐Ÿš€ Title: OCEAN-AI is an open-source app for Big Five personality traits assessment and HR-processes automatization.

๐Ÿค— Demo: ElenaRyumina/OCEANAI

๐Ÿ‘ฅ Authors: @ElenaRyumina , @DmitryRyumin , and Alexey Karpov

๐Ÿ“ Description: OCEAN-AI consists of a set of modules for intellectual analysis of human behavior based on multimodal data for automatic personality traits (PT) assessment. The app evaluates five PT: Openness to experience, Conscientiousness, Extraversion, Agreeableness, Non-Neuroticism.

The App solves practical tasks:
- Ranking of potential candidates by professional responsibilities.
- Forming effective work teams.
- Predicting consumer preferences for industrial goods.

๐Ÿ” Keywords: #OCEANAI #BigFive #PersonalityTraits #PersonalityAnalysis #MultimodalData #Transformers #FirstImpressionsV2 #DeepLearning #Innovation #BehaviorAnalysis #AffectiveRecognition #TeamFormation #ConsumerPreferences #CandidateRanking
  • 1 reply
ยท
posted an update 3 months ago
view post
Post
2294
๐Ÿ•บ๐ŸŽฌ๐Ÿ”ฅ New Research Alert - CVPR 2024 (Avatars Collection)! ๐Ÿ”ฅ๐Ÿค–โšก
๐Ÿ“„ Title: GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh

๐Ÿ“ Description: GoMAvatar is an efficient method for real-time, high-quality, animatable human modeling from a single monocular video. It combines the rendering quality of Gaussian splatting with the geometry modeling capabilities of deformable meshes, enabling realistic digital avatars that can be rearticulated in new poses and rendered from novel angles, while seamlessly integrating with graphics pipelines.

๐Ÿ‘ฅ Authors: Jing Wen, Xiaoming Zhao, Zhongzheng Ren, Alexander G. Schwing, Shenlong Wang

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh (2404.07991)

๐ŸŒ Github Page: https://wenj.github.io/GoMAvatar/
๐Ÿ“ Repository: https://github.com/wenj/GoMAvatar

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #GoMAvatar #3DAvatar #3DAnimation #AnimatableAvatars #MonocularVideo #RealTimeRendering #HumanModeling #CVPR2024 #DeepLearning #Innovation
posted an update 3 months ago
view post
Post
2483
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert (Avatars Collection)! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations ๐Ÿ”

๐Ÿ“ Description: PhysAvatar is a novel framework that uses inverse rendering and physics to autonomously reconstruct the shape, appearance, and physical properties of clothed human avatars from multi-view video data.

๐Ÿ‘ฅ Authors: Yang Zheng et al.

๐Ÿ”— Paper: PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations (2404.04421)

๐ŸŒ GitHub Page: https://qingqing-zhao.github.io/PhysAvatar

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #PhysAvatar #DigitalHumans #InverseRendering #PhysicsSimulation #AvatarModeling #ClothSimulation #PhotorealisticRendering #ComputerVision #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
2242
๐Ÿš€๐Ÿ’ƒ๐ŸŒŸ New Research Alert (Avatars Collection)! ๐ŸŒŸ๐Ÿ•บ๐Ÿš€
๐Ÿ“„ Title: InstructHumans: Editing Animated 3D Human Textures with Instructions

๐Ÿ“ Description: InstructHumans is a novel framework for text-instructed editing of 3D human textures that employs a modified Score Distillation Sampling (SDS-E) method along with spatial smoothness regularization and gradient-based viewpoint sampling to achieve high-quality, consistent, and instruction-true edits.

๐Ÿ‘ฅ Authors: Jiayin Zhu, Linlin Yang, Angela Yao

๐Ÿ”— Paper: InstructHumans: Editing Animated 3D Human Textures with Instructions (2404.04037)

๐ŸŒ Web Page: https://jyzhu.top/instruct-humans
๐Ÿ“ Repository: https://github.com/viridityzhu/InstructHumans

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #InstructHumans #3DTextureEditing #TextInstructions #ScoreDistillationSampling #SDS-E #SpatialSmoothnessRegularization #3DEditing #AvatarEditing #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
2539
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting ๐Ÿ”

๐Ÿ“ Description: 3DGS-Avatar is a novel method for creating animatable human avatars from monocular videos using 3D Gaussian Splatting (3DGS). By using a non-rigid deformation network and as-isometric-as-possible regularizations, the method achieves comparable or better performance than SOTA methods while being 400x faster in training and 250x faster in inference, allowing real-time rendering at 50+ FPS.

๐Ÿ‘ฅ Authors: Zhiyin Qian, Shaofei Wang, Marko Mihajlovic, Andreas Geiger, Siyu Tang

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (2312.09228)

๐ŸŒ Github Page: https://neuralbodies.github.io/3DGS-Avatar/
๐Ÿ“ Repository: https://github.com/mikeqzy/3dgs-avatar-release

๐Ÿ“บ Video: https://www.youtube.com/watch?v=FJ29U9OkmmU

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #3DGSAvatar #3DAvatar #3DGaussianSplatting #AnimatableAvatars #MonocularVideo #RealTimeRendering #FastTraining #EfficientInference #CVPR2024 #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
2276
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Image ๐Ÿ”

๐Ÿ“ Description: GeneAvatar is a generic approach for editing 3D head avatars based on a single 2D image, applicable to different volumetric representations. The novel expression-aware generative modification model delivers high quality and consistent editing results across multiple viewpoints and emotions.

๐Ÿ‘ฅ Authors: Chong Bao et al.

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Image (2404.02152)

๐ŸŒ Github Page: https://zju3dv.github.io/geneavatar/
๐Ÿ“ Repository: https://github.com/zju3dv/GeneAvatar

๐Ÿ“บ Video: https://www.youtube.com/watch?v=4zfbfPivtVU

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #GeneAvatar #HeadAvatar #3DHeadAvatarEditing #VolumetricHeadAvatar #SingleImageEditing #ExpressionAwareModification #CVPR2024 #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
1794
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024 (Avatars Collection)! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: MonoAvatar++: Efficient 3D Implicit Head Avatar with Mesh-anchored Hash Table Blendshapes ๐Ÿ”

๐Ÿ“ Description: MonoAvatar++ is a real-time neural implicit 3D head avatar model with high quality and fine-grained control over facial expressions. It uses local hash table blendshapes attached to a parametric facial model for efficient rendering, achieving SOTA results even for challenging expressions.

๐Ÿ‘ฅ Authors: Ziqian Bai, Feitong Tan, Sean Fanello, Rohit Pandey, Mingsong Dou, Shichen Liu, Ping Tan, Yinda Zhang

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: Efficient 3D Implicit Head Avatar with Mesh-anchored Hash Table Blendshapes (2404.01543)

๐ŸŒ Github Page: https://augmentedperception.github.io/monoavatar-plus

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #MonoAvatar++ #HeadAvatar #3DModeling #AvatarGeneration #NeuralImplicitAvatar #EfficientRendering #CVPR2024 #DeepLearning #Innovation
replied to their post 4 months ago
view reply

Hi @researcher171473,

The idea of using GANs or latent diffusion models to augment visual data instead of image mixing is indeed interesting. However, I have a few considerations:

  1. Training GANs and diffusion models is typically more resource intensive than simple image mixing.
  2. Ensuring that the generated examples are sufficiently informative and diverse to improve the classifier may require additional mechanisms (diversity regularization, adversarial training, latent space manipulation, domain-specific constraints, etc.).
  3. The generated examples must retain their original semantics and class membership to effectively complement the training data.
  4. The classifier may overfit the generated examples and lose performance on real data.

Despite these potential challenges, combining image blending with generative models could potentially yield better results. For example, GANs could be used to generate additional realistic samples that can then be mixed to increase diversity.

posted an update 4 months ago
view post
Post
1413
๐ŸŽฏ๐Ÿ–ผ๏ธ๐ŸŒŸ New Research Alert - ICLR 2024! ๐ŸŒŸ ๐Ÿ–ผ๏ธ๐ŸŽฏ
๐Ÿ“„ Title: Adversarial AutoMixup ๐Ÿ–ผ๏ธ

๐Ÿ“ Description: Adversarial AutoMixup is an approach to image classification augmentation. By alternately optimizing a classifier and a mixed-sample generator, it attempts to generate challenging samples and improve the robustness of the classifier against overfitting.

๐Ÿ‘ฅ Authors: Huafeng Qin et al.

๐Ÿ“… Conference: ICLR, May 7-11, 2024 | Vienna, Austria ๐Ÿ‡ฆ๐Ÿ‡น

๐Ÿ”— Paper: Adversarial AutoMixup (2312.11954)

๐Ÿ“ Repository: https://github.com/JinXins/Adversarial-AutoMixup

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #AutoMixup #ImageClassification #ImageAugmentation #AdversarialLearning #ICLR2024 #DeepLearning #Innovation
  • 4 replies
ยท
posted an update 4 months ago
view post
Post
2133
โ˜๏ธโ˜” New Research Alert! โ„๏ธ๐ŸŒ™
๐Ÿ“„ Title: CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning

๐Ÿ“ Description: CoDA is a UDA methodology that boosts models to understand all adverse scenes (โ˜๏ธ,โ˜”,โ„๏ธ,๐ŸŒ™) by highlighting the discrepancies within these scenes. CoDA achieves state-of-the-art performances on widely used benchmarks.

๐Ÿ‘ฅ Authors: Ziyang Gong, Fuhao Li, Yupeng Deng, Deblina Bhattacharjee, Xiangwei Zhu, Zhenming Ji

๐Ÿ”— Paper: CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning (2403.17369)

๐Ÿ“ Repository: https://github.com/Cuzyoung/CoDA

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #CoDA #DomainAdaptation #VisualPromptTuning #SAVPT #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
1740
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation ๐Ÿ”

๐Ÿ“ Description: AniPortrait is a novel framework for generating photorealistic portrait animations driven by audio and a reference image, with superior facial naturalness, pose variety, and visual quality, with potential applications in facial motion editing and facial reenactment.

๐Ÿ‘ฅ Authors: Huawei Wei, @ZJYang , Zhisheng Wang

๐Ÿ”— Paper: AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation (2403.17694)

๐Ÿ“ Repository: https://github.com/Zejun-Yang/AniPortrait

๐Ÿค— Demo: ZJYang/AniPortrait_official
๐Ÿ”ฅ Model ๐Ÿค–: ZJYang/AniPortrait

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #AniPortrait #Animation #AudioDriven #Photorealistic #FacialAnimation #DeepLearning #Innovation
  • 2 replies
ยท
posted an update 4 months ago
view post
Post
1592
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: FlashFace: Human Image Personalization with High-fidelity Identity Preservation ๐Ÿ”

๐Ÿ“ Description: FlashFace is a personalized photo editing tool that focuses on high-fidelity identity preservation and improved compliance through advanced encoding and integration strategies.

๐Ÿ‘ฅ Authors: Shilong Zhang, Lianghua Huang, @xichenhku et al.

๐Ÿ”— Paper: FlashFace: Human Image Personalization with High-fidelity Identity Preservation (2403.17008)

๐ŸŒ Github Page: https://jshilong.github.io/flashface-page

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #FlashFace #Personalization #HighFidelityIdentity #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
2661
๐Ÿš€๐Ÿ’ƒ๐ŸŒŸ New Research Alert - ICASSP 2024! ๐ŸŒŸ ๐Ÿ•บ๐Ÿš€
๐Ÿ“„ Title: Text2Avatar: Text to 3D Human Avatar Generation with Codebook-Driven Body Controllable Attribute ๐ŸŒŸ

๐Ÿ“ Description: Text2Avatar is a novel approach that can generate realistic 3D human avatars directly from textual descriptions, enabling multi-attribute control and realistic styling, overcoming the challenges of feature coupling and data scarcity in this domain.

๐Ÿ‘ฅ Authors: Chaoqun Gong et al.

๐Ÿ“… Conference: ICASSP, 14-19 April 2024 | Seoul, Korea ๐Ÿ‡ฐ๐Ÿ‡ท

๐Ÿ”— Paper: Text2Avatar: Text to 3D Human Avatar Generation with Codebook-Driven Body Controllable Attribute (2401.00711)

๐ŸŒ Github Page: https://iecqgong.github.io/text2avatar/

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ“ Added to the ICASSP-2023-24-Papers: https://github.com/DmitryRyumin/ICASSP-2023-24-Papers

๐Ÿ” Keywords: #AvatarGeneration #Text2Avatar #ICASSP2024 #DeepLearning #Innovation
  • 1 reply
ยท
posted an update 4 months ago
view post
Post
1831
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians ๐Ÿ”

๐Ÿ“ Description: GaussianAvatars proposes a novel method for creating photorealistic and fully controllable head avatars by combining a parametric morphable face model with a dynamic 3D representation based on rigged 3D Gaussian splats, enabling high-quality rendering and precise animation control.

๐Ÿ‘ฅ Authors: Shenhan Qian, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Simon Giebenhain, Matthias NieรŸner

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (2312.02069)

๐ŸŒ Github Page: https://shenhanqian.github.io/gaussian-avatars
๐Ÿ“ Repository: https://github.com/ShenhanQian/GaussianAvatars

๐Ÿ“บ Video: https://www.youtube.com/watch?v=lVEY78RwU_I

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #HeadAvatar #GaussianAvatars #DynamicGaussians #3DModeling #AvatarGeneration #CVPR2024 #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
1393
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians ๐Ÿ”

๐Ÿ“ Description: Gaussian Head Avatar is a method for generating highly detailed 3D head avatars using dynamic Gaussian functions controlled by a neural network, ensuring ultra-high quality visualization even under limited viewpoints.

๐Ÿ‘ฅ Authors: Yuelang Xu, @ben55 , Zhe Li, @HongwenZhang , @wanglz14 , Zerong Zheng, and @YebinLiu

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians (2312.03029)

๐ŸŒ Github Page: https://yuelangx.github.io/gaussianheadavatar
๐Ÿ“ Repository: https://github.com/YuelangX/Gaussian-Head-Avatar

๐Ÿ“บ Video: https://www.youtube.com/watch?v=kvrrI3EoM5g

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #HeadAvatar #DynamicGaussians #3DModeling #AvatarGeneration #CVPR2024 #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
1854
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - ICLR 2024! ๐ŸŒŸ ๐ŸŽญ๐Ÿš€
๐Ÿ“„ Title: InstructPix2NeRF: Instructed 3D Portrait Editing from a Single Image ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: InstructPix2NeRF is a novel approach to instructed 3D portrait editing from a single image, using a conditional latent 3D diffusion process and a token position randomization strategy to enable multi-semantic editing while preserving the identity of the portrait.

๐Ÿ‘ฅ Authors: Jianhui Li et al.

๐Ÿ“… Conference: ICLR, May 7-11, 2024 | Vienna, Austria ๐Ÿ‡ฆ๐Ÿ‡น

๐Ÿ”— Paper: InstructPix2NeRF: Instructed 3D Portrait Editing from a Single Image (2311.02826)

๐ŸŒ Github Page: https://mybabyyh.github.io/InstructPix2NeRF
๐Ÿ“ Repository: https://github.com/mybabyyh/InstructPix2NeRF

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #InstructPix2NeRF #AvatarCustomization #3DPortrait #DiffusionProcess #IdentityConsistency #ICLR2024 #DeepLearning #Innovation
posted an update 4 months ago
view post
Post
1922
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ ๐Ÿ’ƒ๐Ÿป๐Ÿš€
๐Ÿ“„ Title: NECA: Neural Customizable Human Avatar ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: The NECA paper presents a novel method for creating customizable human avatars from video, allowing detailed manipulation of pose, shadow, shape, lighting, and texture for realistic rendering and editing.

๐Ÿ‘ฅ Authors: Junjin Xiao, Qing Zhang, Zhan Xu, and Wei-Shi Zheng

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: NECA: Neural Customizable Human Avatar (2403.10335)

๐Ÿ“ Repository: https://github.com/iSEE-Laboratory/NECA

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #NECA #AvatarCustomization #RealisticRendering #HumanRepresentation #CVPR2024 #DeepLearning #Animation #Innovation
posted an update 4 months ago
view post
Post
๐Ÿš€๐Ÿ’ƒ๐Ÿป๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿ•บ ๐Ÿš€
๐Ÿ“„ Title: Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: Animatable Gaussians - a novel method for creating lifelike human avatars from RGB videos, utilizing 2D CNNs and 3D Gaussian splatting to capture pose-dependent garment details and dynamic appearances with high fidelity.

๐Ÿ‘ฅ Authors: Zhe Li, Zerong Zheng, Lizhen Wang, and Yebin Liu

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling (2311.16096)

๐ŸŒ Github Page: https://animatable-gaussians.github.io
๐Ÿ“ Repository: https://github.com/lizhe00/AnimatableGaussians

๐Ÿ“บ Video: https://www.youtube.com/watch?v=kOmZxD0HxZI

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #AnimatableGaussians #HumanAvatars #3DGaussianSplatting #CVPR2024 #DeepLearning #Animation #Innovation
posted an update 4 months ago
view post
Post
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert! ๐ŸŒŸ๐ŸŽญ ๐Ÿš€
๐Ÿ“„ Title: VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: VLOGGER is a method for text- and audio-driven generation of talking human video from a single input image of a person, building on the success of recent generative diffusion models.

๐Ÿ‘ฅ Authors: @enriccorona , @Andreiz , @kolotouros , @thiemoall , and et al.

๐Ÿ”— Paper: VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis (2403.08764)

๐ŸŒ Github Page: https://enriccorona.github.io/vlogger/

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #VLOGGER #EmbodiedAvatarSynthesis #MultimodalDiffusion #GenerativeDiffusionModels #DeepLearning #Animation #Innovation
posted an update 4 months ago
view post
Post
๐Ÿš€๐Ÿ—ฃ๏ธ๐ŸŒŸ New Research Alert - ICASSP 2024! ๐ŸŒŸ๐Ÿ—ฃ๏ธ๐Ÿš€
๐Ÿ“„ Title: AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: Diffused Resynthesis and HuBERT Speech Quality Enhancement.

๐Ÿ‘ฅ Authors: Ju-Chieh Chou, Chung-Ming Chien, Karen Livescu

๐Ÿ“… Conference: ICASSP, 14-19 April 2024 | Seoul, Korea ๐Ÿ‡ฐ๐Ÿ‡ท

๐Ÿ”— Paper: AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement (2309.08030)

๐ŸŒ Web Page: https://home.ttic.edu/~jcchou/demo/avse/avse_demo.html

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the
DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Speech Enhancement Collection: DmitryRyumin/speech-enhancement-65de31e1b6d9a040c151702e

๐Ÿ” Keywords: #AV2Wav #SpeechEnhancement #SpeechProcessing #AudioVisual #Diffusion #ICASSP2024 #Innovation
posted an update 4 months ago
view post
Post
๐Ÿš€๐Ÿ•บ๐ŸŒŸ New Research Alert - AAAI 2024! ๐ŸŒŸ๐Ÿ’ƒ๐Ÿš€
๐Ÿ“„ Title: Relightable and Animatable Neural Avatars from Videos ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: Relightable & animatable neural avatars from sparse videos.

๐Ÿ‘ฅ Authors: Wenbin Lin, Chengwei Zheng, Jun-Hai Yong, and Feng Xu

๐Ÿ“… Conference: AAAI, February 20-27, 2024 | Vancouver, Canada ๐Ÿ‡จ๐Ÿ‡ฆ

๐Ÿ”— Paper: Relightable and Animatable Neural Avatars from Videos (2312.12877)

๐ŸŒ Github Page: https://wenbin-lin.github.io/RelightableAvatar-page
๐Ÿ“ Repository: https://github.com/wenbin-lin/RelightableAvatar

๐Ÿ“บ Video: https://www.youtube.com/watch?v=v9rlys0xQGo

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ“š Added to the AAAI 2024 Papers: https://github.com/DmitryRyumin/AAAI-2024-Papers

๐Ÿ” Keywords: #NeuralAvatar #RelightableAvatars #AnimatableAvatars #3DModeling #PhotorealisticRendering #ShadowModeling #DigitalAvatars #GeometryModeling #AAAI2024 #DeepLearning #Animation #Innovation
posted an update 4 months ago
view post
Post
๐Ÿš€๐Ÿ–ผ๏ธ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿ–ผ๏ธ๐Ÿš€
๐Ÿ“„ Title: CAMixerSR: Only Details Need More "Attention" ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: CAMixerSR is a new approach integrating content-aware accelerating framework and token mixer design, to pursue more efficient SR inference via assigning convolution for simple regions but window-attention for complex textures. It exhibits excellent generality and attains competitive results among state-of-the-art models with better complexity-performance trade-offs on large-image SR, lightweight SR, and omnidirectional-image SR.

๐Ÿ‘ฅ Authors: Yan Wang, Shijie Zhao, Yi Liu, Junlin Li, and Li Zhang

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: CAMixerSR: Only Details Need More "Attention" (2402.19289)

๐Ÿ”— Repository: https://github.com/icandle/CAMixerSR

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Image Enhancement Collection: DmitryRyumin/image-enhancement-65ee1cd2fe1c0c877ae55d28

๐Ÿ” Keywords: #CAMixerSR #SuperResolution #WindowAttention #ImageEnhancement #CVPR2024 #DeepLearning #Innovation
posted an update 5 months ago
view post
Post
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - ICLR 2024! ๐ŸŒŸ๐ŸŽญ ๐Ÿš€
๐Ÿ“„ Title: GPAvatar: Generalizable and Precise Head Avatar from Image(s) ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: GPAvatar's objective is to faithfully replicate head avatars while providing precise control over expressions and postures.

๐Ÿ‘ฅ Authors: Xuangeng Chu et al.

๐Ÿ“… Conference: ICLR, May 7-11, 2024 | Vienna, Austria ๐Ÿ‡ฆ๐Ÿ‡น

๐Ÿ”— Paper: GPAvatar: Generalizable and Precise Head Avatar from Image(s) (2401.10215)

๐Ÿ”— Github Page: https://xg-chu.github.io/project_gpavatar
๐Ÿ”— Repository: https://github.com/xg-chu/GPAvatar

๐Ÿ”— Video: https://www.youtube.com/watch?v=7A3DMaB6Zk0

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #GPAvatar #MTA #Synthesis #LipSyncing #Expressions #HighResolutionVideos #ICLR2024 #DeepLearning #Animation #Innovation
  • 1 reply
ยท
posted an update 5 months ago
view post
Post
๐Ÿš€๐ŸŽญ๐ŸŒŸ New Research Alert - ICLR 2024! ๐ŸŒŸ๐ŸŽญ ๐Ÿš€
๐Ÿ“„ Real3D-Portrait: One-shot Realistic 3D Talking Portrait Synthesis ๐ŸŒŸ๐Ÿš€

๐Ÿ‘ฅ Authors: Zhenhui Ye et al.

๐Ÿ“… Conference: ICLR, May 7-11, 2024 | Vienna, Austria ๐Ÿ‡ฆ๐Ÿ‡น

๐Ÿ”— Paper: Real3D-Portrait: One-shot Realistic 3D Talking Portrait Synthesis (2401.08503)

๐Ÿ”— Github Page: https://real3dportrait.github.io/
๐Ÿ”— Repository: https://github.com/yerfor/Real3DPortrait

๐Ÿ”ฅ Model ๐Ÿค–: ameerazam08/Real3DPortrait

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #Real3D-Potrait #I2P #HTB-SR #A2M #Synthesis #LipSyncing #HighResolutionVideos #ICLR2024 #DeepLearning #Animation #Innovation
posted an update 5 months ago
view post
Post
๐Ÿš€๐Ÿ˜ˆ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿ˜ˆ ๐Ÿš€
๐Ÿ“„ Title: SyncTalk: The Devil ๐Ÿ˜ˆ is in the Synchronization for Talking Head Synthesis ๐ŸŒŸ๐Ÿš€

๐Ÿ“ Description: SyncTalk synthesizes synchronized talking head videos, employing tri-plane hash representations to maintain subject identity. It can generate synchronized lip movements, facial expressions, and stable head poses, and restores hair details to create high-resolution videos.

๐Ÿ‘ฅ Authors: Ziqiao Peng et al.

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis (2311.17590)

๐Ÿ”— Github Page: https://ziqiaopeng.github.io/synctalk
๐Ÿ”— Repository: https://github.com/ZiqiaoPeng/SyncTalk

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #TalkingHeads #Synthesis #TriPlaneHash #FacialExpressions #LipSyncing #HighResolutionVideos #CVPR2024 #DeepLearning #Animation #Innovation
posted an update 5 months ago
view post
Post
๐Ÿš€๐ŸŽฌ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐ŸŽฌ ๐Ÿš€
๐Ÿ“„ Title: GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians ๐ŸŒŸ๐Ÿš€

๐Ÿ‘ฅ Authors: Liangxiao Hu et al.

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿ”— Paper: GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (2312.02134)

๐Ÿ”— Github Page: https://huliangxiao.github.io/GaussianAvatar
๐Ÿ”— Repository: https://github.com/huliangxiao/GaussianAvatar

๐Ÿ”— Video: https://www.youtube.com/watch?v=a4g8Z9nCF-k

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #GaussianAvatar #3DGaussians #HumanAvatarModeling #PoseDependentAppearance #DynamicAppearanceModeling #MotionEstimation #MonocularSettings #AppearanceQuality #RenderingEfficiency #CVPR2024 #DeepLearning #Animation #Innovation
posted an update 5 months ago
view post
Post
๐Ÿš€๐Ÿ’ƒ๐ŸŒŸ New Research Alert - CVPR 2024! ๐ŸŒŸ๐Ÿ•บ ๐Ÿš€
๐Ÿ“„ Title: MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model ๐ŸŒŸ๐Ÿš€

๐Ÿ‘ฅ Authors: @junhao910323 , @hansyan et al.

๐Ÿ“… Conference: CVPR, Jun 17-21, 2024 | Seattle WA, USA ๐Ÿ‡บ๐Ÿ‡ธ

๐Ÿค— Demo: zcxu-eric/magicanimate

๐Ÿ”— Paper: MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model (2311.16498)
๐Ÿ”— Github Page: https://showlab.github.io/magicanimate/
๐Ÿ”— Repository: https://github.com/magic-research/magic-animate

๐Ÿ”ฅ Model ๐Ÿค–: zcxu-eric/MagicAnimate

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

๐Ÿ” Keywords: #MagicAnimate #DiffusionModel #HumanImageAnimation #CVPR2024 #Diffusion #DeepLearning #Innovation
posted an update 5 months ago
view post
Post
๐ŸŒŸ๐ŸŽญโœจ Exciting News! The Latest in Expressive Video Portrait Generation! ๐ŸŒŸ๐ŸŽญโœจ

๐Ÿ“„ Title: EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions

๐Ÿ‘ฅ Authors: Linrui Tian, @lucaskingjade , Bang Zhang, and @Liefeng

๐Ÿ”— Paper: EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions (2402.17485)
๐Ÿ”— Github Page: https://humanaigc.github.io/emote-portrait-alive
๐Ÿ”— Repository: https://github.com/HumanAIGC/EMO

๐Ÿ” Keywords: #EMO #EmotePortrait #Audio2VideoDiffusion #ExpressiveAnimations #VideoGeneration #DigitalArt #HumanExpression #ComputerVision #DeepLearning #AI

๐Ÿš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36
posted an update 5 months ago
view post
Post
๐Ÿš€๐Ÿ”ฅ๐ŸŒŸ New Research Alert - ICLR 2024! ๐ŸŒŸ๐Ÿ”ฅ๐Ÿš€
๐Ÿ“„ Title: FuseChat: Revolutionizing Chat Models Fusion ๐ŸŒŸ๐Ÿš€

๐Ÿ‘ฅ Authors: @Wanfq , @passerqxj et al.

๐Ÿ“… Conference: ICLR, May 7-11, 2024 | Vienna, Austria ๐Ÿ‡ฆ๐Ÿ‡น

๐Ÿ”— Paper: FuseChat: Knowledge Fusion of Chat Models (2402.16107)
๐Ÿ”— Repository: https://github.com/fanqiwan/FuseLLM

๐Ÿ”ฅ Models ๐Ÿค–:
1๏ธโƒฃ FuseChat-7B-VaRM: FuseAI/FuseChat-7B-VaRM
2๏ธโƒฃ FuseChat-7B-Slerp: FuseAI/FuseChat-7B-Slerp
3๏ธโƒฃ OpenChat-3.5-7B-Solar: FuseAI/OpenChat-3.5-7B-Solar
4๏ธโƒฃ FuseChat-7B-TA: FuseAI/FuseChat-7B-TA
5๏ธโƒฃ OpenChat-3.5-7B-Mixtral: FuseAI/OpenChat-3.5-7B-Mixtral

๐Ÿ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

๐Ÿ” Keywords: #FuseChat #ChatModels #KnowledgeFusion #ICLR2024 #AI #Innovation #FuseLLM
replied to their post 5 months ago
view reply

Got it, added links to models on the HF hub. Will keep that in mind for the future. ๐Ÿ˜Š