音频驱动视频[ICCV 2025] Official Pytorch Implementation of FLOAT: Generative Motion Latent Flow Matching for Audio-driven Talking Portrait.
active 2025-03-13 → 2026-06-07 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 452 days from 2025-03-13 to 2026-06-07. Pushes: 4 total, peak 4 in a day. Pull requests: 3 total, peak 1 in a day. Issues: 18 total, peak 2 in a day. Comments: 41 total, peak 7 in a day. Stars: 198 total, peak 16 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| TaekyungKi | 22 | 4 | 0 | 13 |
| set-soft | 8 | 0 | 0 | 8 |
| why986 | 5 | 0 | 0 | 5 |
| Egoago | 3 | 0 | 0 | 2 |
| yangw8 | 3 | 0 | 0 | 0 |
| jupinter | 3 | 0 | 0 | 3 |
| reques | 3 | 0 | 0 | 1 |
| adkkml | 2 | 0 | 0 | 2 |
| Ede1994 | 2 | 0 | 0 | 2 |
| nitinmukesh | 2 | 0 | 0 | 2 |
| pavanpankajtruviz | 1 | 0 | 0 | 1 |
| TheNetherWatcher | 1 | 0 | 1 | 0 |
| ChanHyeok-Choi | 1 | 0 | 0 | 0 |
| mohamed-em2m | 1 | 0 | 1 | 0 |
| 18445864529 | 1 | 0 | 0 | 0 |
| SuperMaximus1984 | 1 | 0 | 0 | 1 |
| charmingchi1 | 1 | 0 | 0 | 0 |
| dhb52 | 1 | 0 | 0 | 0 |
| leong-focus | 1 | 0 | 0 | 1 |
| wzr0108 | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#30mohamed-em2m2025-11-26 03:01
- Issue comment#21leong-focus2025-11-13 01:59the question of train float?
- Issue#28ChanHyeok-Choi2025-10-23 12:42Question about conditioning on 3DMM head pose parameters
- Issue#26TaekyungKi2025-09-23 04:56why using w_r_s as part of the condition
- Issue#25TaekyungKi2025-09-23 04:56--no_crop flag seems not working
- Issue comment#25TaekyungKi2025-09-23 04:56--no_crop flag seems not working
- Issue comment#26TaekyungKi2025-09-23 04:53why using w_r_s as part of the condition
- Issue#26184458645292025-09-10 10:04why using w_r_s as part of the condition
- Issue#25dhb522025-09-07 02:31--no_crop flag seems not working
- Issue comment#13SuperMaximus19842025-08-10 07:22Can it support segmented processing of long audio or audio streaming?
- Issue#22TaekyungKi2025-08-03 06:51Relax the license.
- Issue comment#23TaekyungKi2025-08-03 06:48Additional controls would be nice
- Issue#23JKWenstad2025-07-23 23:34Additional controls would be nice
- Issue comment#19why9862025-07-02 05:03Questions about training objective
- Issue comment#19why9862025-07-01 08:18Questions about training objective
- Issue#14TaekyungKi2025-07-01 07:44The inference characters look a bit like foreigners. How can we make them look like Chinese? When will the training code be provided?
- Issue comment#17TaekyungKi2025-07-01 07:44ComfyUI nodes that shows the architecture flow
- Issue comment#19TaekyungKi2025-07-01 07:41Questions about training objective
- Issue#10reques2025-06-27 05:26A question about training
- Issue comment#11why9862025-06-20 09:47help
- Issue comment#11Ede19942025-06-20 06:58help
- Pull request#20ketyi2025-06-19 15:40
- Issue comment#19Egoago2025-06-19 09:50Questions about training objective
- Issue comment#19why9862025-06-19 08:50Questions about training objective
- Issue comment#19Egoago2025-06-19 08:06Questions about training objective
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 198 stars here means stars gained during the window, not the repo's star count.