1st Place Solution of Multiview Egocentric Hand Tracking Challenge ECCV2024
This work addresses hand tracking for VR interaction, but it is incremental as it builds on existing datasets and methods.
The paper tackled multi-view egocentric hand tracking for VR interaction by proposing a method with data augmentation and neural smoothing, achieving 13.92mm MPJPE on Umetrack and 21.66mm MPJPE on HOT3D datasets.
Multi-view egocentric hand tracking is a challenging task and plays a critical role in VR interaction. In this report, we present a method that uses multi-view input images and camera extrinsic parameters to estimate both hand shape and pose. To reduce overfitting to the camera layout, we apply crop jittering and extrinsic parameter noise augmentation. Additionally, we propose an offline neural smoothing post-processing method to further improve the accuracy of hand position and pose. Our method achieves 13.92mm MPJPE on the Umetrack dataset and 21.66mm MPJPE on the HOT3D dataset.