Reconstructing hand-held objects from monocular RGB images is an appealing yet challenging task. In this task, contacts between hands and objects provide important cues for recovering the 3D geometry of the hand-held objects. Though recent works have employed implicit functions to achieve impressive progress, they ignore formulating contacts in their frameworks, which results in producing less realistic object meshes. In this work, we explore how to model contacts in an explicit way to benefit the implicit reconstruction of hand-held objects. The proposed method CHOI consists of two components: explicit contact prediction and implicit shape reconstruction. In the first part, we propose a new subtask of directly estimating 3D hand-object contacts from a single image. The part-level and vertex-level graph-based transformers are cascaded and jointly learned in a coarse-to-fine manner for more accurate contact probabilities. In the second part, we introduce a novel method to diffuse estimated contact states from the hand mesh surface to nearby 3D space and leverage diffused contact probabilities to construct the implicit neural representation for the manipulated object. Benefiting from estimating the interaction patterns between the hand and the object, CHOI can reconstruct more realistic object meshes, especially for object parts that are in contact with hands. Extensive experiments on challenging benchmarks show that the proposed method outperforms the current state of the arts by a great margin.

[Paper] [Code]

Fig 1. The overview of learning explicit contact for implicit reconstruction. In the first stage, the method estimates contact regions of the hand given a monocular RGB image. Based on the template MANO mesh, part- and vertex-level graph-based transformers are constructed and cascaded together for accurate predictions. In the second stage, the estimated contact is used to construct the implicit neural representation. An off-the-shelf module is first utilized to produce the camera parameters, hand mesh, and initial features from the image. Then, the structured contact codes are generated by anchoring contact probabilities to the hand mesh surface. After sparse convolutions, the contact states on the hand surface are diffused to its nearby 3D space, which facilitates the perception and reconstruction of the manipulated object.