Commit Graph

32 Commits

Author SHA1 Message Date
ShowLo 4f69a9cfdd Fix the bug that causes an infinite loop when the total number of frames in the video does not exceed 11.
eg, the video has 11 frames, when select the NO.6 frame, `while abs(random_element - img_idx) <= 5:` will result in an infinite loop
2024-09-19 17:09:35 +08:00
czk32611 98f0e6f2b1 Fixed bug in train.py where pe was missing 2024-08-08 14:56:25 +08:00
czk32611 1de8261491 Merge pull request #85 from shounakb1/train_codes
initial data script
2024-08-06 18:49:07 +08:00
Shounak Banerjee b968548131 fixed mltiple video data preperation 2024-06-17 18:39:15 +00:00
Shounak Banerjee af82f3b00f temporary commit to save changes 2024-06-13 14:14:52 +00:00
Shounak Banerjee d74c4c098b clean code and sepaarate finetuned_inference.py 2024-06-07 18:39:24 +00:00
Shounak Banerjee b4a592d7f3 modified dataloader.py and inference.py for training and inference 2024-06-03 11:09:12 +00:00
czk32611 6d19f3c0c8 Remove crop_audio_window from DataLoader.py 2024-06-01 22:23:47 +08:00
shounak 7254ca6306 initial data script 2024-05-16 18:24:44 +00:00
czk32611 30dcd5237f Update train_codes/README.md 2024-04-30 15:10:03 +08:00
czk32611 d73daf1808 Update draft training codes 2024-04-28 18:04:22 +08:00
czk32611 6e32247cb1 <BUGFIX>: fix app.py (#47) 2024-04-27 14:57:13 +08:00
czk32611 865a68c60e <enhance>: support using float16 in inference to speed up 2024-04-27 14:55:54 +08:00
czk32611 2c52de01b4 Update README and LICENSE 2024-04-26 09:24:23 +08:00
czk32611 ef36109b06 <enhance>(inference): support using an image as video input(#17 #34) 2024-04-19 09:18:39 +08:00
itechmusic f5579a95bd Fix fps calculation bug in realtime_inference.py (#35) 2024-04-18 18:03:58 +08:00
czk32611 0387c39a93 Add codes for real time inference 2024-04-18 12:05:22 +08:00
Oli_Zhan 955ca416ea Merge pull request #31 from tobycroft/main 2024-04-17 14:34:35 +08:00
gluttony-10 d611968ba3 Update requirements.txt 2024-04-16 22:36:04 +08:00
Tuuz ae956e63e0 Update app.py
add force_download to make env build more stable
2024-04-16 19:32:25 +08:00
itechmusic 696ec5aa03 Update README.md 2024-04-16 15:45:01 +08:00
czk32611 4ab40470f7 Add gradio 2024-04-16 09:57:12 +08:00
itechmusic a8a028690a Merge pull request #5 from hotea/main
fix some cuda related issue when run on M-Series Mac with cpu
2024-04-07 15:21:36 +08:00
itechmusic a5114f1fcb Update README.md 2024-04-07 09:49:29 +08:00
= 9a5212c8dd fix some cuda related issue when run on M-Series Mac 2024-04-05 22:03:28 +08:00
zkangchen bc1379abad <enhance>: modified inference codes
1. Can set bbox_shift in configs/inference/test.yaml
2. Do not need to pip install whisper now
2024-04-03 14:56:17 +08:00
czk32611 dde2ee49ef Update demo 2024-04-03 12:32:16 +08:00
itechmusic 6b5fdf2d76 Update README.md 2024-04-03 12:30:36 +08:00
itechmusic 578325be30 Update README.md 2024-04-03 10:57:23 +08:00
itechmusic 4c9c634fc4 Update README.md 2024-04-03 10:44:25 +08:00
phighting 4bb0398b53 Update README.md 2024-04-02 18:04:57 +08:00
czk32611 d861498ae3 initial_commit 2024-04-02 17:56:58 +08:00