Real3DPortrait
Real3DPortrait copied to clipboard
yerfor
Real3D-Portrait: One-shot Realistic 3D Talking Portrait Synthesis; ICLR 2024 Spotlight; Official code
Hello, I would like to ask how can I load the pre-trained model and fine-tune it on my self-collected dataset?
When i ran the CMD: CUDA_VISIBLE_DEVICES=0 python tasks/run.py --config=egs/os_avatar/audio_lm3d_syncnet.yaml --exp_name=audio_lm3d_syncnet --reset. i am getting StopIteration Error and when i checked the dataloader using len(dataloader) the length is 0, i dont...
I encountered a error when doing inference. The following code raise that error: Real3DPortrait/./inference/edit_secc.py, line 91, in blink_eye_for_secc dists, _ = nbrs.kneighbors(face_xys) # [512*512, 1] distance to nearest non-bg pixel
I can do lip sync for any character. I reduced the quality to 10 MB in order to upload video. If you are interested, write to me in telegram: The_best_result...
the provided google collab notebook seems to get stuck on  is this normal for 20+ mins on this line of code? thanks
Hi, I want to know how to adjust the rendering frame rate from 25 to 20 when rendering videos so that I can save time? By the way, is there...
python inference/real3d_infer.py --src_img data/raw/examples/Macron.png --drv_aud data/raw/examples/Obama_5s.wav --drv_pose data/raw/examples/May_5s.mp4 --bg_img data/raw/examples/bg.png --out_name output.mp4 --out_mode concat_debug | WARN: checkpoints/240210_real3dportrait_orig/th1kh_512_audio2motion/base.yaml not exist. | load 'model' from 'checkpoints/240210_real3dportrait_orig/audio2secc_vae/model_ckpt_steps_400000.ckpt', strict=True | WARN: checkpoints/240210_real3dportrait_orig/secc2plane_torso_orig/secc_img2plane_orig.yaml not exist....
the dataset downloaded has no audio track. So I downloaded audio stream audioStream = yt.streams.filter(audio_codec='mp4a.40.2', type="audio").first() then merge the video and audio track via ffmpeg cmd. I tryed download and...
