Cross-Visibility-Gap Identity Persistence for 4D Scene Graphs
4D scene graphs require persistent object identity when an object disappears inside an opaque carrier and later reappears as a fresh detection.
ProofPaper ↗
Key points
- When appearance and target-centric motion are unavailable, identity recovery must rely on relational history.
- We formulate cross-visibility-gap identity persistence as historical-node recovery through a remembered target--carrier binding.
- We introduce CVG-SGM, a perception-frontend-compatible transfer--maintain--read memory.
- Geometry-weighted cross-node interaction transfers target-specific evidence into carrier states, a shared per-node GRU maintains this information during transport, and a two-time readout combines stored carrier--identity compatibility with current detection--carrier geometry to recover the historical identity.
Sources (1)
- [1]Cross-Visibility-Gap Identity Persistence for 4D Scene GraphsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 02:20 AM
4D scene graphs require persistent object identity when an object disappears inside an opaque carrier and later reappears as a fresh detection.
When appearance and target-centric motion are unavailable, identity recovery must rely on relational history.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 19, 2026Introducing Clairic, a new System One class of extraction models
- Oct 19, 2026MiniMax M3.1 Flash shows China's AI price race has no floor in sight
- Oct 18, 2026From Seeing to Reporting: A Theory of Reliable Visual Set Generation in Vision-Language Models
- Oct 18, 2026Google Open-Sources Harness2, a Recursive Agent Harnessing Framework
- Oct 18, 2026MOSS-Image comes to Fizgig
- Oct 18, 2026Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark: DeepSeek-V4.1-Flash-1017 EXL3 2.9bpw on 2x NVIDIA DGX Spark (GB10) with TensorFold: 2.1-2.3x the vLLM kit's decode speed.