wx44wx/icassp2027-experiment
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add no-box full-frame Qwen AV baselines

Browse Source
This commit is contained in:
Xin Wang
2026-09-17 02:51:26 +08:00
parent a9f748d6a6
commit 62134a0fa0
10 changed files with 1164 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
PROJECT_CONTEXT.md
View File

@@ -81,6 +81,12 @@ All are pending in that log; Omni LoRA is deferred and no semantic gate is compl
### Implemented but not GPU-verified
- Two no-box full-frame Qwen AV screens are implemented on top of the matched
global-Omni + recent-Qwen-audio inputs. F2 queries the latest strictly
pre-cutoff post-merger spatial-token grid; F3 first applies a shared causal
GRU to the latest same-grid-position histories. They use no face boxes or
person IDs and therefore test full-frame visual context, not speaker
grounding. See [`docs/easycom_fullframe_qwen_av.md`](docs/easycom_fullframe_qwen_av.md).
- Qwen2.5-Omni now has two matched chunk-handling controls around the existing
5 Hz partial-tail path: native completed 2-second blocks held between
boundaries, and non-native 200-ms payload commits. Both preserve dense causal