ECCV 2026 Paper EPO Refines 3D Foundation Model Poses Without Feature Tracks
rsasaki0109 · x · 2026-10-03
- rsasaki0109 shares the official repo of EPO (Edge-Based Pose Optimization), an ECCV 2026 paper by Mattia D'Urso et al.
- EPO is an optimization framework that improves Structure-from-Motion reconstructions from 3D foundation models (e.g., VGGT) via edge-based gradient-descent pose refinement, without explicit feature tracks.
- Visualizations on the Graz Town Hall scene (TerraSky3D) show poses evolving from VGGT's initial estimate to refined results closely matching ground truth.
- Code is open source (111 stars on GitHub) with demo scripts and a notebook.
More from Research
- Blogger ships RL tutorial after an all-nighter; next CUDA post due Friday — goyal__pramod · 2026-10-03
- AgentBug-Smith paper: top coding agents fix just 9% of agent harness bugs — rohanpaul_ai · 2026-10-03
- CoRL 2026 to host Sim-to-Real-to-Field workshop on transfer science, CFP open — berkeley_ai · 2026-10-03
- HPE Labs review covers photonic AI accelerators with wavelength/temporal-domain GEMM and QD comb lasers — jwt0625 · 2026-10-03
- Hinton: we design the learning algorithm, but we don't really understand how neural networks work — austinc3301 · 2026-10-03
- Policy Gradient for LLMs, Explained Visually: A From-Scratch REINFORCE Derivation — joecole · 2026-10-03