Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces DICArt, a novel framework for category-level articulated object pose estimation that formulates pose estimation as a conditional discrete diffusion process operating in a discrete state-space. DICArt employs a flexible flow decider to dynamically balance denoising and resetting of tokens during diffusion, and incorporates a hierarchical kinematic coupling strategy to respect the object's kinematic structure. Experiments on synthetic and real-world datasets demonstrate DICArt's improved performance and robustness compared to continuous regression methods.
Ditch continuous pose regression for articulated objects: DICArt's discrete diffusion approach unlocks more reliable 6D pose estimation by respecting kinematic constraints and navigating the search space more effectively.
Articulated object pose estimation is a core task in embodied AI. Existing methods typically regress poses in a continuous space, but often struggle with 1) navigating a large, complex search space and 2) failing to incorporate intrinsic kinematic constraints. In this work, we introduce DICArt (DIsCrete Diffusion for Articulation Pose Estimation), a novel framework that formulates pose estimation as a conditional discrete diffusion process. Instead of operating in a continuous domain, DICArt progressively denoises a noisy pose representation through a learned reverse diffusion procedure to recover the GT pose. To improve modeling fidelity, we propose a flexible flow decider that dynamically determines whether each token should be denoised or reset, effectively balancing the real and noise distributions during diffusion. Additionally, we incorporate a hierarchical kinematic coupling strategy, estimating the pose of each rigid part hierarchically to respect the object's kinematic structure. We validate DICArt on both synthetic and real-world datasets. Experimental results demonstrate its superior performance and robustness. By integrating discrete generative modeling with structural priors, DICArt offers a new paradigm for reliable category-level 6D pose estimation in complex environments.