2cfcfdd1af
feat: Phase 2.6 edges migration to Qdrant (TKG-only architecture)
...
Phase 2.6.1: co_occurrence_edges migration
- build_co_occurrence_edges_from_qdrant()
- Qdrant embeddings → frame grouping → YOLO objects
- Result: 6679 edges (vs 6701 PostgreSQL)
Phase 2.6.2: face_face_edges migration
- build_face_face_edges_from_qdrant()
- Qdrant embeddings → frame grouping → face pairs
- mutual_gaze detection preserved
- Result: 6 edges (exact match)
Phase 2.6.3: speaker_face_edges migration
- build_speaker_face_edges_from_qdrant()
- Qdrant embeddings → trace_id frame ranges
- SPEAKS_AS edge creation
Architecture:
- All edges use Qdrant payload (no face_detections queries)
- PostgreSQL fallback for empty Qdrant
- Estimated 3.6x performance improvement
Testing:
- Playground (3003): ✓ All Phase 2.6 logs verified
- Edge counts: ✓ Close match with PostgreSQL
- Fallback: ✓ Working
Docs:
- docs_v1.0/DESIGN/TKG_PHASE2_6_EDGES_MIGRATION.md
- docs_v1.0/M4_workspace/2026-06-21_phase2_6_test.md
2026-06-21 04:47:49 +08:00
17e4e15860
feat: add Vision LLM integration (CLIP + Qwen3-VL cascade)
...
- Add Qwen3-VL dynamic management (start/stop/status CLI)
- Add CLIP + Qwen3-VL cascade detection strategy
- Add Vision CLI commands (vision start/stop/status, detect)
- Add cascade_vision processor module
- Add clip processor module
- Add qwen_vl_manager module
Changes:
- scripts/start_qwen3vl.sh, stop_qwen3vl.sh: Qwen3-VL management scripts
- src/core/vision/: Qwen3-VL manager module
- src/core/processor/cascade_vision.rs: CLIP + Qwen3-VL cascade logic
- src/core/processor/clip.rs: CLIP classification and detection
- src/api/clip_api.rs: CLIP API endpoints
- src/cli/vision.rs: Vision CLI implementation
- src/cli/args.rs: Add Vision and Detect commands
- src/main.rs: Integrate Vision CLI
- src/core/mod.rs: Add vision module
- src/core/processor/mod.rs: Add cascade_vision module
2026-06-13 16:25:52 +08:00
834b0d4865
feat: score-based search, LLM re-ranking endpoint, video title search, pipeline module
...
Core search changes:
- Replace RRF with score-based merge (max of semantic/keyword/identity)
- Add video title ILIKE search for brand/name queries (score 0.9)
- Add /api/v1/search/llm-smart endpoint with Gemma 4 re-ranking
- Fix LLM JSON parsing (markdown fences, empty responses)
Infrastructure:
- Rebuild Qdrant collection (clear 347K contaminated points)
- Add dotenv loading to main.rs for config parity
- Implement store_pre_chunk in postgres_db.rs
Pipeline module (WordPress):
- store-asrx, rule1, vectorize, phase1, complete endpoints
- CLI commands for pipeline operations
Docs:
- SEARCH_SCORE_IMPROVEMENT.md (score-based merge proposal)
2026-06-04 07:40:41 +08:00
874d688987
feat: deploy hybrid search (semantic+keyword+identity) with RRF fusion
...
- Replace smart_search with hybrid RRF implementation
- Add speaker_detections table for identity-agent binding
- Fix identity queries: direct SQL to avoid type mismatches
- Add debug logs to job_worker for processor debugging
- Deployed to production (3002) successfully
Key changes:
- search.rs: Complete rewrite with 3 strategies + RRF
- postgres_db.rs: speaker_detections table + identity query fixes
- job_worker.rs: Debug logs for output file checks
Tested:
- Hybrid search works with semantic + keyword + identity
- Identity search: 'identity:Charade' returns correct results
- Chinese keyword search: '調光' matches Charade summaries
Bugs found:
- Case mismatch: 'ASRX' vs 'asrx' in processors field
- Missing CUT dependency for ASRX processor
2026-06-01 15:15:17 +08:00
3a33d00449
refactor: modularize server.rs into separate route modules
...
- Extract scan.rs, files.rs, types.rs, processing.rs, visual_chunk_search.rs
- Move AppState and AppConfig to types.rs
- Each module exposes pub fn xxx_routes() -> Router<AppState>
- server.rs reduced from 5005 to 118 lines (orchestrator only)
- All stubs filled with real implementations from git history
- Verify: cargo check, clippy, tests all pass
2026-05-21 16:38:49 +08:00
362c63007c
feat: smart search response includes start_frame/end_frame/fps, add limit param
2026-05-18 01:21:43 +08:00
4125163f7b
refactor: rename search uuid -> file_uuid
2026-05-18 01:17:48 +08:00
3164a65554
update: pipeline, search, clip, embedding fixes
2026-05-17 19:46:35 +08:00
39ba5ddf76
feat: Phase 1 handover - schema migration, correction mechanism, API fixes
...
Schema changes: dev.chunks->dev.chunk, remove old_chunk_id/chunk_index
Correction: asr-1.json format, generate/apply scripts
API: 37/37 endpoints fixed and tested
Docs: HANDOVER_V2.0.md for M4
2026-05-11 07:03:22 +08:00
736b14be15
Fix smart search: use EmbeddingGemma instead of Ollama mxbai
2026-05-08 01:23:34 +08:00
573714788f
Release v1.0.0 candidate
2026-05-08 00:48:15 +08:00
74b6182eba
feat: media API (video/bbox/thumbnail), UUID unification, dot matrix text, portal fixes, API dictionary V1.3
2026-05-06 13:34:49 +08:00
2b23d1cfbd
feat: update core API, database layer, and worker modules
...
- Remove unused imports (n8n_search, universal_search, Client, Arc, etc.)
- Update API endpoints for identity, face recognition, search
- Fix postgres_db.rs search_videos parent_uuid column
- Add snapshot API and identity agent API
- Clean up backup files (.bak, .bak2)
2026-04-30 15:07:02 +08:00
1f84e5469f
feat: backup architecture docs, source code, and scripts
2026-04-25 17:15:45 +08:00
Accusys