EventEgoHands++: Event-based Egocentric 3D Hand Mesh Reconstruction with Real Dataset
EventEgoHands++ reconstructs egocentric 3D hand meshes from event cameras using instance-level detection and a 1M-frame real dataset.
EventEgoHands++ adds a Hand Detector estimating instance-level bounding boxes and masks for both hands, plus Adaptive Attention that dynamically gates attention based on detection results to learn inter-hand relationships. The authors extend the synthetic N-HOT3D dataset and construct EEH-R, the largest real-world event-based egocentric hand dataset to date, with roughly 1 million annotated frames including low-light conditions. Experiments on synthetic and real datasets show consistent improvements over baselines.
- Instance-level hand detection distinguishes left and right hands
- Adaptive Attention gates attention using detection results
- EEH-R dataset contains ~1M annotated real-world event-camera egocentric frames
- Captures include low-light conditions
- Outperforms baselines on both synthetic and real benchmarks
- Products
- EventEgoHands++N-HOT3DEEH-R
From the original article
1 min read at huggingface.co
3D hand mesh reconstruction is a challenging yet essential task for downstream applications, including human-robot interaction and AR/VR. Although conventional cameras have been widely adopted for this task, methods that rely on them struggle in low-light environments and under severe motion blur.
By Ryosei Hara, Wataru Ikeda, Masashi Hatano · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.