Multi-view Image-based Hand Geometry Refinement using Differentiable Monte Carlo Ray Tracing
This work addresses annotation quality issues in hand pose estimation datasets, which is important for researchers in computer vision and robotics, though it is incremental as it builds on existing datasets and methods.
The paper tackled the problem of inaccurate annotations in hand pose and shape estimation datasets by proposing a refinement approach using differentiable ray tracing, which improved annotation quality on the InterHand2.6M dataset and showed significant improvements in synthetic and real data evaluations.
The amount and quality of datasets and tools available in the research field of hand pose and shape estimation act as evidence to the significant progress that has been made.However, even the datasets of the highest quality, reported to date, have shortcomings in annotation. We propose a refinement approach, based on differentiable ray tracing,and demonstrate how a high-quality publicly available, multi-camera dataset of hands(InterHand2.6M) can become an even better dataset, with respect to annotation quality. Differentiable ray tracing has not been employed so far to relevant problems and is hereby shown to be superior to the approximative alternatives that have been employed in the past. To tackle the lack of reliable ground truth, as far as quantitative evaluation is concerned, we resort to realistic synthetic data, to show that the improvement we induce is indeed significant. The same becomes evident in real data through visual evaluation.