Understanding Everyday Hands in Action from RGB-D Images

Gregory Rogez (1) , James Steven Supancic (2) , Deva Ramanan (3)

1 LEAR - Learning and recognition in vision
2 UCI - University of Californie, Irvine
3 CMU - Carnegie Mellon University [Pittsburgh]

We analyze functional manipulations of handheld objects, formalizing the problem as one of fine-grained grasp classification. To do so, we make use of a recently developed fine-grained taxonomy of human-object grasps. We introduce a large dataset of 12000 RGB-D images covering 71 everyday grasps in natural interactions. Our dataset is different from past work (typically addressed from a robotics perspective) in terms of its scale, diversity, and combination of RGB and depth data. From a computer-vision perspective , our dataset allows for exploration of contact and force prediction (crucial concepts in functional grasp analysis) from perceptual cues. We present extensive experimental results with state-of-the-art baselines, illustrating the role of segmentation, object context, and 3D-understanding in functional grasp analysis. We demonstrate a near 2X improvement over prior work and a naive deep baseline, while pointing out important directions for improvement.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Gregory Rogez, James Steven Supancic, Deva Ramanan. Understanding Everyday Hands in Action from RGB-D Images. ICCV - IEEE International Conference on Computer Vision, Dec 2015, Santiago, Chile. pp.3889-3897, ⟨10.1109/ICCV.2015.443⟩. ⟨hal-01237011⟩