Institution profile

Disney Research

Industry researchnorthamerica · us
Official website
Research library35linked papers
Opportunities0open roles
Selected work

Representative Papers

Design and Control of a Bipedal Robotic Character

Jul 15, 2024Robotics

To address the limited expressiveness and poor terrain adaptability of bipedal robots in entertainment applications, this paper proposes a dynamic control framework tailored for humanoid stage performances. Methodologically: (1) it introduces a character-driven mechanical design that jointly optimizes artistic expressivity and locomotion robustness; (2) it develops an action-conditioned reinforcement learning controller enabling real-time synthesis and blending of multi-source motion animations; and (3) it integrates online dynamic gait planning with an intuitive human–robot interaction interface. Experimental results demonstrate stable locomotion over complex terrains and high-fidelity, low-latency stage performances with enhanced expressivity. The system significantly improves affective human–robot connection and audience immersion. This work establishes a novel paradigm for entertainment robotics that unifies artistic expression with adaptive motor intelligence.

4 citationsRead paper

Interactive Generative Motion Editing via Scheduled Inpainting

Jul 31, 2026

Existing motion editing methods struggle to simultaneously achieve large-scale structural modifications and faithful preservation of original motion, while generative models often lack interactive editing capabilities. This work proposes scheduled inpainting—a novel approach that dynamically controls the balance between preserving and generating motion in specific spatiotemporal regions during inference of a generative model. For the first time, this method unifies generative motion synthesis with interactive editing. It enables flexible operations such as extension, stitching, and composition while maintaining natural motion quality, and achieves high-precision editing through fine-grained spatiotemporal control. Experiments demonstrate that the proposed method outperforms four baselines across multiple tasks, and both ablation studies and user evaluations confirm its effectiveness and practical utility.

0 citationsRead paper

BFMTrack: Latent Sequence Optimization for Physics-Based Motion Tracking with Behavioral Foundation Models

Jun 23, 2026

Existing behavioral foundation models (BFMs) struggle to accurately track time-varying targets—such as complex motion sequences—due to the absence of temporal dynamics in their latent spaces. This work proposes Latent Sequence Optimization (LSO), a method that directly optimizes temporally coherent latent trajectories within the BFM latent space. By integrating physics-based simulation rollouts with policy gradient updates, LSO achieves precise motion tracking without requiring handcrafted reward functions. The approach innovatively incorporates temporally correlated noise modeling, substantially enhancing trajectory smoothness and detail fidelity, thereby overcoming the limitation of BFMs to time-invariant tasks. Experiments demonstrate that LSO enables high-fidelity, highly generalizable motion reproduction across diverse scenarios, including dense trajectory tracking, sparse keyframe control, and real-world deployment on humanoid robots.

0 citationsRead paper
Recent publications

Latest Papers

Interactive Generative Motion Editing via Scheduled Inpainting

Jul 31, 2026

Existing motion editing methods struggle to simultaneously achieve large-scale structural modifications and faithful preservation of original motion, while generative models often lack interactive editing capabilities. This work proposes scheduled inpainting—a novel approach that dynamically controls the balance between preserving and generating motion in specific spatiotemporal regions during inference of a generative model. For the first time, this method unifies generative motion synthesis with interactive editing. It enables flexible operations such as extension, stitching, and composition while maintaining natural motion quality, and achieves high-precision editing through fine-grained spatiotemporal control. Experiments demonstrate that the proposed method outperforms four baselines across multiple tasks, and both ablation studies and user evaluations confirm its effectiveness and practical utility.

0 citationsRead paper

BFMTrack: Latent Sequence Optimization for Physics-Based Motion Tracking with Behavioral Foundation Models

Jun 23, 2026

Existing behavioral foundation models (BFMs) struggle to accurately track time-varying targets—such as complex motion sequences—due to the absence of temporal dynamics in their latent spaces. This work proposes Latent Sequence Optimization (LSO), a method that directly optimizes temporally coherent latent trajectories within the BFM latent space. By integrating physics-based simulation rollouts with policy gradient updates, LSO achieves precise motion tracking without requiring handcrafted reward functions. The approach innovatively incorporates temporally correlated noise modeling, substantially enhancing trajectory smoothness and detail fidelity, thereby overcoming the limitation of BFMs to time-invariant tasks. Experiments demonstrate that LSO enables high-fidelity, highly generalizable motion reproduction across diverse scenarios, including dense trajectory tracking, sparse keyframe control, and real-world deployment on humanoid robots.

0 citationsRead paper

αDepth: Learning Single-Pass Soft Boundary Decomposition for Stereo Conversion

May 29, 2026

This work addresses the challenge of accurately modeling soft boundaries—such as hair or motion blur—in stereo conversion, where foreground-background blending leads to ambiguous depth correspondences that conventional single-layer depth estimation fails to capture. To overcome this limitation, the authors propose αDepth, a method that jointly estimates color and depth through a layered representation in a single forward pass. Central to their approach is the novel Cyclic Alpha Representation (CAR), which reformulates soft-boundary modeling from a global object-centric task into a local boundary decomposition process. This enables unsupervised, high-quality layered inference even in multi-object scenes. The method effectively mitigates background color bleeding and structural distortions at soft boundaries, significantly enhancing visual fidelity in stereo conversion tasks.

0 citationsRead paper