Arrow Research search

Author name cluster

Kaiwen Chen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
2 author rows

Possible papers

2

NeurIPS Conference 2025 Conference Paper

Towards Dynamic 3D Reconstruction of Hand-Instrument Interaction in Ophthalmic Surgery

  • Ming Hu
  • Zhengdi Yu
  • Feilong Tang
  • Kaiwen Chen
  • Yulong Li
  • Imran Razzak
  • Junjun He
  • Tolga Birdal

Accurate 3D reconstruction of hands and instruments is critical for vision-based analysis of ophthalmic microsurgery, yet progress has been hampered by the lack of realistic, large-scale datasets and reliable annotation tools. In this work, we introduce OphNet-3D, the first extensive RGB-D dynamic 3D reconstruction dataset for ophthalmic surgery, comprising 41 sequences from 40 surgeons and totaling 7. 1 million frames, with fine-grained annotations of 12 surgical phases, 10 instrument categories, dense MANO hand meshes, and full 6-DoF instrument poses. To scalably produce high-fidelity labels, we design a multi-stage automatic annotation pipeline that integrates multi-view data observation, data-driven motion prior with cross-view geometric consistency and biomechanical constraints, along with a combination of collision-aware interaction constraints for instrument interactions. Building upon OphNet-3D, we establish two challenging benchmarks—bimanual hand pose estimation and hand–instrument interaction reconstruction—and propose two dedicated architectures: H-Net for dual-hand mesh recovery and OH-Net for joint reconstruction of two-hand–two-instrument interactions. These models leverage a novel spatial reasoning module with weak-perspective camera modeling and collision-aware center-based representation. Both architectures outperform existing methods by substantial margins, achieving improvements of over 2mm in Mean Per Joint Position Error (MPJPE) and up to 23\% in ADD-S metrics for hand and instrument reconstruction, respectively.

PDF Details

IROS Conference 2018 Conference Paper

A Topology-Based Path Similarity Metric and its Application to Sampling-Based Motion Planning

  • Jory Denny
  • Kaiwen Chen
  • Hanglin Zhou

Many applications of robotic motion planning benefit from considering multiple homotopically distinct paths rather than a single path from start to goal. However, determining whether paths represent different homotopy classes can be difficult to compute. We propose metrics for efficiently approximating the homotopic similarity of two paths are, instead of verifying homotopy equivalence directly. We propose two metrics: (1) a naive application of local planning, a common subroutine of sampling-based motion planning, and (2) a novel approach that reasons about the topologically distinct portions of the workspace that a path visits. We present three applications of our metric to demonstrate its use and effectiveness: extracting topologically distinct paths from an existing roadmap, comparing paths for robot manipulators, and improving the computational efficiency of an existing sampling-based method, Path Deformation Roadmaps (PDRs), by over two orders of magnitude. We explore the trade-off between quality and computational efficiency in the proposed metrics.

Details