Upload 58 files

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/S3DIS/Area_1_conferenceRoom_1.txt filter=lfs diff=lfs merge=lfs -text
+data/S3DIS/Area_2_WC_1.txt filter=lfs diff=lfs merge=lfs -text
+data/S3DIS/Area_4_lobby_2.txt filter=lfs diff=lfs merge=lfs -text
+data/S3DIS/Area_5_office_3.txt filter=lfs diff=lfs merge=lfs -text
+data/S3DIS/Area_6_office_9.txt filter=lfs diff=lfs merge=lfs -text

checkpoints/download_ckpts.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+
+# Define the URLs for the checkpoints
+BASE_URL="https://dl.fbaipublicfiles.com/segment_anything_2/072824/"
+sam2_hiera_t_url="${BASE_URL}sam2_hiera_tiny.pt"
+sam2_hiera_s_url="${BASE_URL}sam2_hiera_small.pt"
+sam2_hiera_b_plus_url="${BASE_URL}sam2_hiera_base_plus.pt"
+sam2_hiera_l_url="${BASE_URL}sam2_hiera_large.pt"
+
+
+# Download each of the four checkpoints using wget
+echo "Downloading sam2_hiera_tiny.pt checkpoint..."
+wget $sam2_hiera_t_url || { echo "Failed to download checkpoint from $sam2_hiera_t_url"; exit 1; }
+
+echo "Downloading sam2_hiera_small.pt checkpoint..."
+wget $sam2_hiera_s_url || { echo "Failed to download checkpoint from $sam2_hiera_s_url"; exit 1; }
+
+echo "Downloading sam2_hiera_base_plus.pt checkpoint..."
+wget $sam2_hiera_b_plus_url || { echo "Failed to download checkpoint from $sam2_hiera_b_plus_url"; exit 1; }
+
+echo "Downloading sam2_hiera_large.pt checkpoint..."
+wget $sam2_hiera_l_url || { echo "Failed to download checkpoint from $sam2_hiera_l_url"; exit 1; }
+
+echo "All checkpoints are downloaded successfully."

checkpoints/sam2_hiera_l.yaml

@@ -0,0 +1,117 @@
+# @package _global_
+
+# Model
+model:
+  _target_: sam2.modeling.sam2_base.SAM2Base
+  image_encoder:
+    _target_: sam2.modeling.backbones.image_encoder.ImageEncoder
+    scalp: 1
+    trunk:
+      _target_: sam2.modeling.backbones.hieradet.Hiera
+      embed_dim: 144
+      num_heads: 2
+      stages: [2, 6, 36, 4]
+      global_att_blocks: [23, 33, 43]
+      window_pos_embed_bkg_spatial_size: [7, 7]
+      window_spec: [8, 4, 16, 8]
+    neck:
+      _target_: sam2.modeling.backbones.image_encoder.FpnNeck
+      position_encoding:
+        _target_: sam2.modeling.position_encoding.PositionEmbeddingSine
+        num_pos_feats: 256
+        normalize: true
+        scale: null
+        temperature: 10000
+      d_model: 256
+      backbone_channel_list: [1152, 576, 288, 144]
+      fpn_top_down_levels: [2, 3]  # output level 0 and 1 directly use the backbone features
+      fpn_interp_model: nearest
+
+  memory_attention:
+    _target_: sam2.modeling.memory_attention.MemoryAttention
+    d_model: 256
+    pos_enc_at_input: true
+    layer:
+      _target_: sam2.modeling.memory_attention.MemoryAttentionLayer
+      activation: relu
+      dim_feedforward: 2048
+      dropout: 0.1
+      pos_enc_at_attn: false
+      self_attention:
+        _target_: sam2.modeling.sam.transformer.RoPEAttention
+        rope_theta: 10000.0
+        feat_sizes: [32, 32]
+        embedding_dim: 256
+        num_heads: 1
+        downsample_rate: 1
+        dropout: 0.1
+      d_model: 256
+      pos_enc_at_cross_attn_keys: true
+      pos_enc_at_cross_attn_queries: false
+      cross_attention:
+        _target_: sam2.modeling.sam.transformer.RoPEAttention
+        rope_theta: 10000.0
+        feat_sizes: [32, 32]
+        rope_k_repeat: True
+        embedding_dim: 256
+        num_heads: 1
+        downsample_rate: 1
+        dropout: 0.1
+        kv_in_dim: 64
+    num_layers: 4
+
+  memory_encoder:
+      _target_: sam2.modeling.memory_encoder.MemoryEncoder
+      out_dim: 64
+      position_encoding:
+        _target_: sam2.modeling.position_encoding.PositionEmbeddingSine
+        num_pos_feats: 64
+        normalize: true
+        scale: null
+        temperature: 10000
+      mask_downsampler:
+        _target_: sam2.modeling.memory_encoder.MaskDownSampler
+        kernel_size: 3
+        stride: 2
+        padding: 1
+      fuser:
+        _target_: sam2.modeling.memory_encoder.Fuser
+        layer:
+          _target_: sam2.modeling.memory_encoder.CXBlock
+          dim: 256
+          kernel_size: 7
+          padding: 3
+          layer_scale_init_value: 1e-6
+          use_dwconv: True  # depth-wise convs
+        num_layers: 2
+
+  num_maskmem: 7
+  image_size: 1024
+  # apply scaled sigmoid on mask logits for memory encoder, and directly feed input mask as output mask
+  sigmoid_scale_for_mem_enc: 20.0
+  sigmoid_bias_for_mem_enc: -10.0
+  use_mask_input_as_output_without_sam: true
+  # Memory
+  directly_add_no_mem_embed: true
+  # use high-resolution feature map in the SAM mask decoder
+  use_high_res_features_in_sam: true
+  # output 3 masks on the first click on initial conditioning frames
+  multimask_output_in_sam: true
+  # SAM heads
+  iou_prediction_use_sigmoid: True
+  # cross-attend to object pointers from other frames (based on SAM output tokens) in the encoder
+  use_obj_ptrs_in_encoder: true
+  add_tpos_enc_to_obj_ptrs: false
+  only_obj_ptrs_in_the_past_for_eval: true
+  # object occlusion prediction
+  pred_obj_scores: true
+  pred_obj_scores_mlp: true
+  fixed_no_obj_ptr: true
+  # multimask tracking settings
+  multimask_output_for_tracking: true
+  use_multimask_token_for_obj_ptr: true
+  multimask_min_pt_num: 0
+  multimask_max_pt_num: 1
+  use_mlp_for_obj_ptr_proj: true
+  # Compilation flag
+  compile_image_encoder: False

checkpoints/sam2_hiera_large.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7442e4e9b732a508f80e141e7c2913437a3610ee0c77381a66658c3a445df87b
+size 897952466

data/KITTI/scene1.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25c46ce954b520e299998d736075ab95f6d34c130ef22a9e2ec479625091e096
+size 2946128

data/KITTI/scene2.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75e2fed53105a16182f718181d4ec4aa9331829acb5b58ca6d4701b2274df624
+size 3019136

data/KITTI/scene3.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fbaa7bafedaa0d2c23bd1b252ff2e4c8a42cc51917f496e78b78954c107f378
+size 2945984

data/KITTI/scene4.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8de268a14bcf2bb405a13118ad259e96ec712e56dc451db447ff71a9b429ffc0
+size 2854472

data/KITTI/scene5.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:459f6439ec9396b2f9d3877e08ded8258e79959067b073e80d75c989322de208
+size 2747624

data/KITTI/scene6.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac870cc73427d338a880439c4192458fb3919b8845f9a805e9d4157b8ecb4cca
+size 2904032

data/Objaverse/blender.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12271fc3bdd3f7ca32a6ec1bb5db3cb600924470fda04725643e3645e4efbca7
+size 196736

data/Objaverse/bowl.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02b676f9d17d67c56d27d8dd4d03719425b4110320acd500241df1f4f9e0a858
+size 196736

data/Objaverse/bus_shelter.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25b9a0c7904737be30521886c4dc5d28c989e626dc983449abc1eafe4ae7cba9
+size 196736

data/Objaverse/dinner_booth.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10222cdff3bcd08518cea82bd886f83d7f1b41e638cd065148302af822e3516b
+size 196736

data/Objaverse/dragon.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e31d82269f55dcc34c0062b5cb63808dd8edf3f730ad3d48082d73916f7120d9
+size 196736

data/Objaverse/elephant.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c02a270e0e15db71c48b184e48016c71e4b38aed27124ae833f48fe30bf69e7
+size 196736

data/Objaverse/horse.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73ae3a359f88dbec43d4ad279968bcc49c2cce49262e2eabe7521b2c9e0ac924
+size 196736

data/Objaverse/human.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e72a515ed1cf2c6ee91cff6694e941fd6690df6267b38241a2f3ce4074c9cfcd
+size 196736

data/Objaverse/human_face.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ab1999321b39d54f9c32c22cceaec551c7d4d15b40525306cfb0879429a2ff8
+size 196736

data/Objaverse/knife_rest.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24c21061c4bf0be74b2a552678fa2c57113aa61457bf4f84f60c9352c5945fd6
+size 196736

data/Objaverse/lamp.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68e726385c64cb9b6543bc8331ece5c60f140e1f6cf3c9423e26b78119e8c64a
+size 196736

data/Objaverse/lock.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79e1ee95be372611447e7b67514ac628c7eec0fd61c267c8e58f14b443aecf0c
+size 196736

data/Objaverse/ornament.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ae79bf0963e6014431cd2ce19c390054b7c39dd6c6033370c80dcc61038826c
+size 196736

data/Objaverse/pipe.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f720a5650f924e0236f27d9dbc11d7386f4a209c7f7e0f5fd8da2bff7e37b2b2
+size 196736

data/Objaverse/planet.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72c09995a54c5345841a604ce4d7e52b5fe7c75114a1603f2b6854d587062ddd
+size 196736

data/Objaverse/plant.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56395c2d6e2e084f2bdfa0dc2f5b8c44aa3d96e78611c9b2b8ed3b3dcafe805a
+size 196736

data/Objaverse/plant_wrong.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b038298c4ea530f20710d8dd0c7dd660d11549dc2ee859c04b4f8167d80f170
+size 196736

data/Objaverse/popcorn_machine.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:354d4c693ad832b7ff0a195053f0eec1060f656c70cb5d5e0096ac0adee71185
+size 196736

data/Objaverse/sculpture.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:441dd723868527cc52a50240e5e4ebd860f637a75620ad92a905a44e736be170
+size 196736

data/Objaverse/skateboard.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdc53c2faa21695ec8a098fa0f8aeb61fad11b63f4b8517b0f57f0a34a1fc800
+size 196736

data/Objaverse/stove.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b03e5d76d5586282d6234d8ef8b58200e29e921c067d82261f9a52b3305b728
+size 196736

data/Objaverse/table.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38ea3caa90c2dd2c7bc328b6a1494ebe7ae7b5031b5ef236a8823d314c475270
+size 196736

data/Objaverse/telescope.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea96631e833796608909beadf9df612446aa5fdebbe974fa310fec6a16990d42
+size 196736

data/Objaverse/thor_hammer.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ab113491eaf9f3260d5916a27db2cd5c4c67e358a0e1a8ed3982354190348ad
+size 196736

data/Objaverse/truck.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f2f61a64735cd811f3f1e8f0734306392b2bf66fb14121fb3ef1ec6f867898c
+size 196736

data/S3DIS/Area_1_conferenceRoom_1.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39a613d67e7fb38c657973c46766263b67354416ef73e089acd2355ffcabd0e0
+size 36686956

data/S3DIS/Area_2_WC_1.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:977d744070877229807fb6195389d1806fe0de1e3ef16850088d33bb03637c15
+size 18941190

data/S3DIS/Area_4_lobby_2.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12761464763a3fc1b18ea37bb21d1238c4ad0d57b2bc8b71ade1aa616adfbd0d
+size 63795347

data/S3DIS/Area_5_office_3.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b61896a9c64611cdfb961f9f7f18f0f95b45c60f3cbf1807793bee5df4bfb94c
+size 24476159

data/S3DIS/Area_6_office_9.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d51578cbf65c4ec9227af73d98f366c1ef2f4ba0d8b5bf21671f1078ddd651c3
+size 51029073

data/ScanNet/scene0000_00.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48d9c949e240ef790c2c86a0b176ac0742349e3156a87d3653aacd6e6ffe6e31
+size 2481775

data/ScanNet/scene0001_01.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2196a2a2a765fe6db2452cfd2bc39ee2f7709d44ad6adc254454bf0df96db2b3
+size 3876911

data/ScanNet/scene0002_00.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:889ccfe6710fbd0e66f84d25a1ce2d13eeb939d40112359804a345e993ca3692
+size 9176943

data/ScanNet/scene0005_01.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:199a37ee72d207566be1b135aeb740bc2e22a1496474d1105cce24e67cb874d8
+size 3833007

data/ScanNet/scene0010_01.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90f61d5e3786d1f26a8b32ddd66bf9a948e16431b5af7c4414d0e9d56f580e6a
+size 4440047

data/ScanNet/scene0016_02.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69be19ecda8d8db0df7bb73c329c7c803b054759aea0e707571c85fe2ddbcbe8
+size 2988591

data/ScanNet/scene0019_01.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:128966f660cb901a7876b9bf37f1b05a038ed89127ae0a5d43bd86652181e9ba
+size 5029423

data/Semantic3D/new.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc6c6f856779d8858bfab3a04eaec69fbb53a08d7f1f7ec821d9f2ce2d1abd56
+size 48000128

data/Semantic3D/patch0.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1631c6afbc61ef0f2617d00f93331edcf790f6e2c88eafddc1cbd3d0de86f7f9
+size 43200128

data/Semantic3D/patch1.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc6c6f856779d8858bfab3a04eaec69fbb53a08d7f1f7ec821d9f2ce2d1abd56
+size 43200128