{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "True\n", "True\n" ] } ], "source": [ "import torch\n", "\n", "# this ensures that the current MacOS version is at least 12.3+\n", "print(torch.backends.mps.is_available())\n", "# this ensures that the current current PyTorch installation was built with MPS activated.\n", "print(torch.backends.mps.is_built())" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "## ALL DEPENDENCIES \n", "import ipywidgets as widgets\n", "import glob\n", "import matplotlib.pyplot as plt\n", "\n", "import sys\n", "sys.path.append(\"thirdparty/AdaptiveWingLoss\")\n", "import os, glob\n", "import numpy as np\n", "import cv2\n", "import argparse\n", "from src.approaches.train_image_translation import Image_translation_block\n", "import torch\n", "import pickle\n", "import face_alignment\n", "from face_alignment import face_alignment \n", "from src.autovc.AutoVC_mel_Convertor_retrain_version import AutoVC_mel_Convertor\n", "import shutil\n", "import time\n", "import util.utils as util\n", "from scipy.signal import savgol_filter\n", "from src.approaches.train_audio2landmark import Audio2landmark_model" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "\n", "# print(\"Choose the image name to animate: (saved in folder 'MakeItTalk/examples/')\")\n", "# img_list = glob.glob1('examples', '*.jpg')\n", "# img_list.sort()\n", "# img_list = [item.split('.')[0] for item in img_list]\n", "# default_head_name = widgets.Dropdown(options=img_list, value='marlene_v2')\n", "# def on_change(change):\n", "# if change['type'] == 'change' and change['name'] == 'value':\n", "# plt.imshow(plt.imread('MakeItTalk/examples/{}.jpg'.format(default_head_name.value)))\n", "# plt.axis('off')\n", "# plt.show()\n", "# default_head_name.observe(on_change)\n", "# display(default_head_name)\n", "# plt.imshow(plt.imread('MakeItTalk/examples/{}.jpg'.format(default_head_name.value)))\n", "# plt.axis('off')\n", "# plt.show()\n", "\n", "image = 'marlene_v2.jpg'\n", "input_path = f'MakeItTalk/examples/{image}.jpg'" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "#@markdown # Animation Controllers\n", "#@markdown Amplify the lip motion in horizontal direction\n", "AMP_LIP_SHAPE_X = 2 #@param {type:\"slider\", min:0.5, max:5.0, step:0.1}\n", "\n", "#@markdown Amplify the lip motion in vertical direction\n", "AMP_LIP_SHAPE_Y = 2 #@param {type:\"slider\", min:0.5, max:5.0, step:0.1}\n", "\n", "#@markdown Amplify the head pose motion (usually smaller than 1.0, put it to 0. for a static head pose)\n", "AMP_HEAD_POSE_MOTION = 0.35 #@param {type:\"slider\", min:0.0, max:1.0, step:0.05}\n", "\n", "#@markdown Add naive eye blink\n", "ADD_NAIVE_EYE = True #@param [\"False\", \"True\"] {type:\"raw\"}\n", "\n", "#@markdown If your image has an opened mouth, put this as True, else False\n", "CLOSE_INPUT_FACE_MOUTH = True #@param [\"False\", \"True\"] {type:\"raw\"} \n", "\n", "\n", "#@markdown # Landmark Adjustment\n", "\n", "#@markdown Adjust upper lip thickness (postive value means thicker)\n", "UPPER_LIP_ADJUST = -1 #@param {type:\"slider\", min:-3.0, max:3.0, step:1.0}\n", "\n", "#@markdown Adjust lower lip thickness (postive value means thicker)\n", "LOWER_LIP_ADJUST = -1 #@param {type:\"slider\", min:-3.0, max:3.0, step:1.0}\n", "\n", "#@markdown Adjust static lip width (in multipication)\n", "LIP_WIDTH_ADJUST = 1.0 #@param {type:\"slider\", min:0.8, max:1.2, step:0.01}" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "sys.stdout = open(os.devnull, 'a')\n", "\n", "parser = argparse.ArgumentParser()\n", "parser.add_argument('--jpg', type=str, default=image)\n", "parser.add_argument('--close_input_face_mouth', default=CLOSE_INPUT_FACE_MOUTH, action='store_true')\n", "parser.add_argument('--load_AUTOVC_name', type=str, default='MakeItTalk/examples/ckpt/ckpt_autovc.pth')\n", "parser.add_argument('--load_a2l_G_name', type=str, default='MakeItTalk/examples/ckpt/ckpt_speaker_branch.pth')\n", "parser.add_argument('--load_a2l_C_name', type=str, default='MakeItTalk/examples/ckpt/ckpt_content_branch.pth') #ckpt_audio2landmark_c.pth')\n", "parser.add_argument('--load_G_name', type=str, default='MakeItTalk/examples/ckpt/ckpt_116_i2i_comb.pth') #ckpt_image2image.pth') #ckpt_i2i_finetune_150.pth') #c\n", "parser.add_argument('--amp_lip_x', type=float, default=AMP_LIP_SHAPE_X)\n", "parser.add_argument('--amp_lip_y', type=float, default=AMP_LIP_SHAPE_Y)\n", "parser.add_argument('--amp_pos', type=float, default=AMP_HEAD_POSE_MOTION)\n", "parser.add_argument('--reuse_train_emb_list', type=str, nargs='+', default=[]) # ['iWeklsXc0H8']) #['45hn7-LXDX8']) #['E_kmpT-EfOg']) #'iWeklsXc0H8', '29k8RtSUjE0', '45hn7-LXDX8',\n", "parser.add_argument('--add_audio_in', default=False, action='store_true')\n", "parser.add_argument('--comb_fan_awing', default=False, action='store_true')\n", "parser.add_argument('--output_folder', type=str, default='examples')\n", "parser.add_argument('--test_end2end', default=True, action='store_true')\n", "parser.add_argument('--dump_dir', type=str, default='', help='')\n", "parser.add_argument('--pos_dim', default=7, type=int)\n", "parser.add_argument('--use_prior_net', default=True, action='store_true')\n", "parser.add_argument('--transformer_d_model', default=32, type=int)\n", "parser.add_argument('--transformer_N', default=2, type=int)\n", "parser.add_argument('--transformer_heads', default=2, type=int)\n", "parser.add_argument('--spk_emb_enc_size', default=16, type=int)\n", "parser.add_argument('--init_content_encoder', type=str, default='')\n", "parser.add_argument('--lr', type=float, default=1e-3, help='learning rate')\n", "parser.add_argument('--reg_lr', type=float, default=1e-6, help='weight decay')\n", "parser.add_argument('--write', default=False, action='store_true')\n", "parser.add_argument('--segment_batch_size', type=int, default=1, help='batch size')\n", "parser.add_argument('--emb_coef', default=3.0, type=float)\n", "parser.add_argument('--lambda_laplacian_smooth_loss', default=1.0, type=float)\n", "parser.add_argument('--use_11spk_only', default=False, action='store_true')\n", "parser.add_argument('-f')\n", "opt_parser = parser.parse_args()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "img = cv2.imread('MakeItTalk/examples/' + opt_parser.jpg)\n", "plt.imshow(img)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "#get the facial landmarks in the image. Run this on a GPU as it can be slow \n", "predictor = face_alignment.FaceAlignment(face_alignment.LandmarksType._3D, device='mps', flip_input=True)\n", "shapes = predictor.get_landmarks(img)\n", "if (not shapes or len(shapes) != 1):\n", " print('Cannot detect face landmarks. Exit.')\n", " exit(-1)\n", "shape_3d = shapes[0]" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Loaded Image...\n" ] } ], "source": [ "#this block runs if the character's mouth is open\n", "if(opt_parser.close_input_face_mouth):\n", " util.close_input_face_mouth(shape_3d)\n", "\n", "#this makes any adjustments necessary to the facial landmarks based on user input \n", "shape_3d[48:, 0] = (shape_3d[48:, 0] - np.mean(shape_3d[48:, 0])) * LIP_WIDTH_ADJUST + np.mean(shape_3d[48:, 0]) # wider lips\n", "shape_3d[49:54, 1] -= UPPER_LIP_ADJUST # thinner upper lip\n", "shape_3d[55:60, 1] += LOWER_LIP_ADJUST # thinner lower lip\n", "shape_3d[[37,38,43,44], 1] -=2. # larger eyes\n", "shape_3d[[40,41,46,47], 1] +=2. # larger eyes\n", "shape_3d, scale, shift = util.norm_input_face(shape_3d)\n", "\n", "print(\"Loaded Image...\", file=sys.stderr)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/Users/marlenemhangami/miniconda3/lib/python3.9/site-packages/resemblyzer/audio.py:33: FutureWarning: Pass orig_sr=16000, target_sr=16000 as keyword args. From version 0.10 passing these as positional arguments will result in an error\n", " wav = librosa.resample(wav, source_sr, sampling_rate)\n", "/Users/marlenemhangami/miniconda3/lib/python3.9/site-packages/resemblyzer/audio.py:47: FutureWarning: Pass y=[0.00289917 0.00289917 0.00289917 ... 0. 0. 0. ], sr=16000 as keyword args. From version 0.10 passing these as positional arguments will result in an error\n", " frames = librosa.feature.melspectrogram(\n", "/Users/marlenemhangami/Downloads/MakeItTalk-main/src/autovc/retrain_version/vocoder_spec/extract_f0_func.py:97: FutureWarning: Pass sr=16000, n_fft=1024 as keyword args. From version 0.10 passing these as positional arguments will result in an error\n", " mel_basis = mel(16000, 1024, fmin=90, fmax=7600, n_mels=80).T\n", "/Users/marlenemhangami/miniconda3/lib/python3.9/site-packages/resemblyzer/audio.py:47: FutureWarning: Pass y=[0.00286865 0.00286865 0.00286865 ... 0. 0. 0. ], sr=16000 as keyword args. From version 0.10 passing these as positional arguments will result in an error\n", " frames = librosa.feature.melspectrogram(\n", "Loaded audio...\n" ] } ], "source": [ "#now we want to load the audio file \n", "# au_data = []\n", "# au_emb = []\n", "# ains = glob.glob1('examples', '*.wav')\n", "# ains = [item for item in ains if item != 'tmp.wav']\n", "# ains.sort()\n", "\n", "#we want an input .wav file \n", "input_audio = 'yourmoment.wav'\n", "\n", "os.system(f'ffmpeg -y -loglevel error -i MakeItTalk/examples/{input_audio} -ar 16000 MakeItTalk/examples/tmp.wav')\n", "shutil.copyfile('MakeItTalk/examples/tmp.wav', f'MakeItTalk/examples/{input_audio}')\n", "\n", "# au embedding\n", "from thirdparty.resemblyer_util.speaker_emb import get_spk_emb\n", "me, ae = get_spk_emb(f'MakeItTalk/examples/{input_audio}')\n", "au_emb.append(me.reshape(-1))\n", "\n", "c = AutoVC_mel_Convertor('examples')\n", "\n", "au_data_i = c.convert_single_wav_to_autovc_input(audio_filename=input_audio, autovc_model_path=opt_parser.load_AUTOVC_name)\n", "\n", "if(os.path.isfile('MakeItTalk/examples/tmp.wav')):\n", " os.remove('MakeItTalk/examples/tmp.wav')\n", "\n", "print(\"Loaded audio...\", file=sys.stderr)\n", "\n" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "# create a landmark fake placeholder\n", "fl_data = []\n", "rot_tran, rot_quat, anchor_t_shape = [], [], []\n", "for au, info in au_data:\n", " au_length = au.shape[0]\n", " fl = np.zeros(shape=(au_length, 68 * 3))\n", " fl_data.append((fl, info))\n", " rot_tran.append(np.zeros(shape=(au_length, 3, 4)))\n", " rot_quat.append(np.zeros(shape=(au_length, 4)))\n", " anchor_t_shape.append(np.zeros(shape=(au_length, 68 * 3)))\n", "\n", "if(os.path.exists(os.path.join('examples', 'dump', 'random_val_fl.pickle'))):\n", " os.remove(os.path.join('examples', 'dump', 'random_val_fl.pickle'))\n", "if(os.path.exists(os.path.join('examples', 'dump', 'random_val_fl_interp.pickle'))):\n", " os.remove(os.path.join('examples', 'dump', 'random_val_fl_interp.pickle'))\n", "if(os.path.exists(os.path.join('examples', 'dump', 'random_val_au.pickle'))):\n", " os.remove(os.path.join('examples', 'dump', 'random_val_au.pickle'))\n", "if (os.path.exists(os.path.join('examples', 'dump', 'random_val_gaze.pickle'))):\n", " os.remove(os.path.join('examples', 'dump', 'random_val_gaze.pickle'))\n", "\n", "with open(os.path.join('examples', 'dump', 'random_val_fl.pickle'), 'wb') as fp:\n", " pickle.dump(fl_data, fp)\n", "with open(os.path.join('examples', 'dump', 'random_val_au.pickle'), 'wb') as fp:\n", " pickle.dump(au_data, fp)\n", "with open(os.path.join('examples', 'dump', 'random_val_gaze.pickle'), 'wb') as fp:\n", " gaze = {'rot_trans':rot_tran, 'rot_quat':rot_quat, 'anchor_t_shape':anchor_t_shape}\n", " pickle.dump(gaze, fp)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/Users/marlenemhangami/Downloads/MakeItTalk-main/src/approaches/train_audio2landmark.py:98: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", " z = torch.tensor(torch.zeros(aus.shape[0], 128), requires_grad=False, dtype=torch.float).to(device)\n", "OpenCV: FFMPEG: tag 0x47504a4d/'MJPG' is not supported with codec id 7 and format 'mp4 / MP4 (MPEG-4 Part 14)'\n", "OpenCV: FFMPEG: fallback to use tag 0x7634706d/'mp4v'\n", "ffmpeg version 5.1.2 Copyright (c) 2000-2022 the FFmpeg developers\n", " built with Apple clang version 14.0.0 (clang-1400.0.29.202)\n", " configuration: --prefix=/opt/homebrew/Cellar/ffmpeg/5.1.2_4 --enable-shared --enable-pthreads --enable-version3 --cc=clang --host-cflags= --host-ldflags= --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libaribb24 --enable-libbluray --enable-libdav1d --enable-libmp3lame --enable-libopus --enable-librav1e --enable-librist --enable-librubberband --enable-libsnappy --enable-libsrt --enable-libsvtav1 --enable-libtesseract --enable-libtheora --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libspeex --enable-libsoxr --enable-libzmq --enable-libzimg --disable-libjack --disable-indev=jack --enable-videotoolbox --enable-neon\n", " libavutil 57. 28.100 / 57. 28.100\n", " libavcodec 59. 37.100 / 59. 37.100\n", " libavformat 59. 27.100 / 59. 27.100\n", " libavdevice 59. 7.100 / 59. 7.100\n", " libavfilter 8. 44.100 / 8. 44.100\n", " libswscale 6. 7.100 / 6. 7.100\n", " libswresample 4. 7.100 / 4. 7.100\n", " libpostproc 56. 6.100 / 56. 6.100\n", "Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'MakeItTalk/examples/tmp.mp4':\n", " Metadata:\n", " major_brand : isom\n", " minor_version : 512\n", " compatible_brands: isomiso2mp41\n", " encoder : Lavf58.76.100\n", " Duration: 00:00:10.70, start: 0.000000, bitrate: 5876 kb/s\n", " Stream #0:0[0x1](und): Video: mjpeg (Baseline) (mp4v / 0x7634706D), yuvj420p(pc, bt470bg/unknown/unknown), 400x400, 5873 kb/s, 62.50 fps, 62.50 tbr, 10k tbn (default)\n", " Metadata:\n", " handler_name : VideoHandler\n", " vendor_id : [0][0][0][0]\n", "Guessed Channel Layout for Input Stream #1.0 : mono\n", "Input #1, wav, from 'MakeItTalk/examples/marlene_sound.wav':\n", " Duration: 00:00:10.99, bitrate: 256 kb/s\n", " Stream #1:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 16000 Hz, mono, s16, 256 kb/s\n", "Stream mapping:\n", " Stream #0:0 -> #0:0 (mjpeg (native) -> h264 (libx264))\n", " Stream #1:0 -> #0:1 (pcm_s16le (native) -> aac (native))\n", "Press [q] to stop, [?] for help\n", "[libx264 @ 0x130e064e0] using cpu capabilities: ARMv8 NEON\n", "[libx264 @ 0x130e064e0] profile High, level 3.0, 4:2:0, 8-bit\n", "[libx264 @ 0x130e064e0] 264 - core 164 r3095 baee400 - H.264/MPEG-4 AVC codec - Copyleft 2003-2022 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00\n", "Output #0, mp4, to 'MakeItTalk/examples/marlene_sound_av.mp4':\n", " Metadata:\n", " major_brand : isom\n", " minor_version : 512\n", " compatible_brands: isomiso2mp41\n", " encoder : Lavf59.27.100\n", " Stream #0:0(und): Video: h264 (avc1 / 0x31637661), yuvj420p(pc, bt470bg/unknown/unknown, progressive), 400x400, q=2-31, 62.50 fps, 16k tbn (default)\n", " Metadata:\n", " handler_name : VideoHandler\n", " vendor_id : [0][0][0][0]\n", " encoder : Lavc59.37.100 libx264\n", " Side data:\n", " cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: N/A\n", " Stream #0:1: Audio: aac (LC) (mp4a / 0x6134706D), 16000 Hz, mono, fltp, 69 kb/s\n", " Metadata:\n", " encoder : Lavc59.37.100 aac\n", "frame= 669 fps=0.0 q=-1.0 Lsize= 537kB time=00:00:10.75 bitrate= 409.1kbits/s speed=20.9x \n", "video:431kB audio:95kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 2.122605%\n", "[libx264 @ 0x130e064e0] frame I:3 Avg QP:10.29 size: 5368\n", "[libx264 @ 0x130e064e0] frame P:184 Avg QP:24.28 size: 1317\n", "[libx264 @ 0x130e064e0] frame B:482 Avg QP:30.64 size: 378\n", "[libx264 @ 0x130e064e0] consecutive B-frames: 2.8% 2.1% 3.6% 91.5%\n", "[libx264 @ 0x130e064e0] mb I I16..4: 66.7% 20.3% 13.0%\n", "[libx264 @ 0x130e064e0] mb P I16..4: 0.7% 1.6% 0.1% P16..4: 6.2% 3.8% 2.5% 0.0% 0.0% skip:85.1%\n", "[libx264 @ 0x130e064e0] mb B I16..4: 0.3% 0.3% 0.0% B16..8: 9.5% 1.8% 0.9% direct: 0.2% skip:87.1% L0:51.7% L1:46.8% BI: 1.5%\n", "[libx264 @ 0x130e064e0] 8x8 transform intra:50.2% inter:6.8%\n", "[libx264 @ 0x130e064e0] coded y,uvDC,uvAC intra: 3.2% 16.1% 10.5% inter: 1.3% 3.6% 3.3%\n", "[libx264 @ 0x130e064e0] i16 v,h,dc,p: 77% 19% 4% 0%\n", "[libx264 @ 0x130e064e0] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 9% 8% 82% 0% 0% 0% 0% 0% 0%\n", "[libx264 @ 0x130e064e0] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 30% 25% 22% 4% 3% 3% 5% 3% 5%\n", "[libx264 @ 0x130e064e0] i8c dc,h,v,p: 54% 26% 20% 0%\n", "[libx264 @ 0x130e064e0] Weighted P-Frames: Y:4.3% UV:0.0%\n", "[libx264 @ 0x130e064e0] ref P L0: 45.2% 17.6% 18.8% 17.6% 0.8%\n", "[libx264 @ 0x130e064e0] ref B L0: 77.1% 17.0% 6.0%\n", "[libx264 @ 0x130e064e0] ref B L1: 90.5% 9.5%\n", "[libx264 @ 0x130e064e0] kb/s:329.32\n", "[aac @ 0x130e07710] Qavg: 41381.832\n", "Audio->Landmark...\n" ] } ], "source": [ "model = Audio2landmark_model(opt_parser, jpg_shape=shape_3d)\n", "if(len(opt_parser.reuse_train_emb_list) == 0):\n", " model.test(au_emb=au_emb)\n", "else:\n", " model.test(au_emb=None)\n", "\n", "print(\"Audio->Landmark...\", file=sys.stderr)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "OpenCV: FFMPEG: tag 0x67706a6d/'mjpg' is not supported with codec id 7 and format 'mp4 / MP4 (MPEG-4 Part 14)'\n", "OpenCV: FFMPEG: fallback to use tag 0x7634706d/'mp4v'\n", "[W NNPACK.cpp:53] Could not initialize NNPACK! Reason: Unsupported hardware.\n", "1 / 1: Landmark->Face...\n", "Done!\n" ] } ], "source": [ "fls = glob.glob1('examples', 'pred_fls_*.txt')\n", "fls.sort()\n", "\n", "for i in range(0,len(fls)):\n", " fl = np.loadtxt(os.path.join('examples', fls[i])).reshape((-1, 68,3))\n", " print(fls[i])\n", " fl[:, :, 0:2] = -fl[:, :, 0:2]\n", " fl[:, :, 0:2] = fl[:, :, 0:2] / scale - shift\n", "\n", " if (ADD_NAIVE_EYE):\n", " fl = util.add_naive_eye(fl)\n", "\n", " # additional smooth\n", " fl = fl.reshape((-1, 204))\n", " fl[:, :48 * 3] = savgol_filter(fl[:, :48 * 3], 15, 3, axis=0)\n", " fl[:, 48*3:] = savgol_filter(fl[:, 48*3:], 5, 3, axis=0)\n", " fl = fl.reshape((-1, 68, 3))\n", "\n", " ''' STEP 6: Imag2image translation '''\n", " model = Image_translation_block(opt_parser, single_test=True)\n", " with torch.no_grad():\n", " model.single_test(jpg=img, fls=fl, filename=fls[i], prefix=opt_parser.jpg.split('.')[0])\n", " print('finish image2image gen')\n", " os.remove(os.path.join('examples', fls[i]))\n", "\n", " print(\"{} / {}: Landmark->Face...\".format(i+1, len(fls)), file=sys.stderr)\n", "print(\"Done!\", file=sys.stderr)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Generated video from image and sound clip" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from IPython.display import Video\n", "\n", "Video(\"MakeItTalk/examples/marlenes_v1.mp4\")\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Display animation: MakeItTalk/examples/paint_boy_pred_fls_M6_04_16k_audio_embed.mp4\n" ] }, { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from IPython.display import HTML\n", "from base64 import b64encode\n", "\n", "for ain in ains:\n", " OUTPUT_MP4_NAME = '{}_pred_fls_{}_audio_embed.mp4'.format(\n", " opt_parser.jpg.split('.')[0],\n", " ain.split('.')[0]\n", " )\n", " mp4 = open('MakeItTalk/examples/{}'.format(OUTPUT_MP4_NAME),'rb').read()\n", " data_url = \"data:video/mp4;base64,\" + b64encode(mp4).decode()\n", "\n", " print('Display animation: MakeItTalk/examples/{}'.format(OUTPUT_MP4_NAME), file=sys.stderr)\n", " display(HTML(\"\"\"\n", " \n", " \"\"\" % data_url))" ] } ], "metadata": { "interpreter": { "hash": "5c7b89af1651d0b8571dde13640ecdccf7d5a6204171d6ab33e7c296e100e08a" }, "kernelspec": { "display_name": "Python 3.11.1 64-bit", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.5" }, "orig_nbformat": 4 }, "nbformat": 4, "nbformat_minor": 2 }