Research map / 3D vision / Face and shape
EvAC3D: From Event-Based Apparent Contours to 3D Models via Continuous Visual Hulls
Ziyun Wang, Kenneth Chaney, Kostas Daniilidis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| carve per event with ACEs instead of 24-view mask visual hull | visual hull carving · MOEC-3D, ground-truth ACEs (mean) | Chamfer distance: 3.2652 -> 2.4267 | Improved |
| carve per event with ACEs instead of 12-view mask visual hull | visual hull carving · MOEC-3D, ground-truth ACEs (mean) | Chamfer distance: 4.3856 -> 2.4267 | Improved |
| use ACE-based carving instead of 24-view mask carving | visual hull carving · MOEC-3D, ground-truth ACEs (mean) | normal consistency: 0.9377 -> 0.9487 | Improved |
| use EvAC3D with network-predicted ACEs instead of Mask 24 | visual hull carving · MOEC-3D real objects (mean) | Chamfer distance: 3.614 -> 2.551 | Improved |
| use EvAC3D instead of Mask 12 on real objects | visual hull carving · MOEC-3D real objects (mean) | Chamfer distance: 8.312 -> 2.551 | Improved |
| use EvAC3D instead of masks from reconstructed images (24 views) | visual hull carving · MOEC-3D real objects (mean) | Chamfer distance: 3.602 -> 2.551 | Improved |
| use EvAC3D instead of E3D silhouette optimization | event-based reconstruction · MOEC-3D real objects (mean) | Chamfer distance: 8.837 -> 2.551 | Improved |
| use EvAC3D instead of 24-view carving to cut operations | visual hull carving · MOEC-3D real objects (mean) | number of carving operations: 6,661,111 -> 1,921,976 | Improved |
| increase mask views from 12 to 24 (ablation) | mask-based carving · MOEC-3D real objects (mean) | Chamfer distance: 8.312 -> 3.614 | Improved |
| ACEs versus 24-view mask carving on individual objects with ground-truth ACEs | EvAC3D · MOEC-3D per object | Chamfer distance: Jello (b) 5.9083 vs 4.4409; Vitamin 2.6190 vs 1.4478 | Mixed |
Closest papers by meaning
- EvDiff3D: Event-Aware Diffusion Repair for High-Fidelity Event-Based 3D Reconstruction (AAAI 2026 · similarity 0.88)
- EA3D: Event-Augmented 3D Diffusion for Generalizable Novel View Synthesis (ICLR 2026 · similarity 0.87)
- Event-3DGS: Event-based 3D Reconstruction Using 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.86)
- The Shape of Events: Edge-Based Inductive Biases via Cross-Domain Distillation (NeurIPS 2026 · similarity 0.86)
- EAG3R: Event-Augmented 3D Geometry Estimation for Dynamic and Extreme-Lighting Scenes (NeurIPS 2025 · similarity 0.85)
- EvGGS: A Collaborative Learning Framework for Event-based Generalizable Gaussian Splatting (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
