Fast SAM 3D Body: Accelerating SAM 3D Body for Real-Time Full-Body Human Mesh Recovery

TL;DR

Fast SAM 3D Body significantly accelerates 3D human mesh recovery from monocular images, enabling real-time applications without sacrificing accuracy by reformulating inference and replacing iterative processes with direct mappings.

Contribution

We introduce a training-free acceleration framework that reformulates inference, enabling real-time 3D human mesh recovery and joint kinematics extraction from monocular images.

Findings

Achieves up to 10.9x end-to-end speedup.

Maintains comparable or better reconstruction fidelity.

Enables real-time humanoid control in teleoperation systems.

Abstract

SAM 3D Body (3DB) achieves state-of-the-art accuracy in monocular 3D human mesh recovery, yet its inference latency of several seconds per image precludes real-time application. We present Fast SAM 3D Body, a training-free acceleration framework that reformulates the 3DB inference pathway to achieve interactive rates. By decoupling serial spatial dependencies and applying architecture-aware pruning, we enable parallelized multi-crop feature extraction and streamlined transformer decoding. Moreover, to extract the joint-level kinematics (SMPL) compatible with existing humanoid control and policy learning frameworks, we replace the iterative mesh fitting with a direct feedforward mapping, accelerating this specific conversion by over 10,000x. Overall, our framework delivers up to a 10.9x end-to-end speedup while maintaining on-par reconstruction fidelity, even surpassing 3DB on benchmarks such as LSPET. We demonstrate its utility by deploying Fast SAM 3D Body in a vision-only teleoperation system that-unlike methods reliant on wearable IMUs-enables real-time humanoid control and the direct collection of manipulation policies from a single RGB stream.