Pith. sign in

REVIEW 5 cited by

DPoser: Diffusion Model as Robust 3D Human Pose Prior

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2312.05541 v2 pith:EKIH57YD submitted 2023-12-09 cs.CV

Junzhe Lu , Jing Lin , Hongkun Dou , Ailing Zeng , Yue Deng , Yulun Zhang , Haoqian Wang This is my paper
classification cs.CV
keywords posedposerhumandiffusionpriorrobusttasksacross
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This work targets to construct a robust human pose prior. However, it remains a persistent challenge due to biomechanical constraints and diverse human movements. Traditional priors like VAEs and NDFs often exhibit shortcomings in realism and generalization, notably with unseen noisy poses. To address these issues, we introduce DPoser, a robust and versatile human pose prior built upon diffusion models. DPoser regards various pose-centric tasks as inverse problems and employs variational diffusion sampling for efficient solving. Accordingly, designed with optimization frameworks, DPoser seamlessly benefits human mesh recovery, pose generation, pose completion, and motion denoising tasks. Furthermore, due to the disparity between the articulated poses and structured images, we propose truncated timestep scheduling to enhance the effectiveness of DPoser. Our approach demonstrates considerable enhancements over common uniform scheduling used in image domains, boasting improvements of 5.4%, 17.2%, and 3.8% across human mesh recovery, pose completion, and motion denoising, respectively. Comprehensive experiments demonstrate the superiority of DPoser over existing state-of-the-art pose priors across multiple tasks.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. PHD: Personalized 3D Human Body Fitting with Point Diffusion

    cs.CV 2025-08 conditional novelty 7.0 of 10

    PHD personalizes 3D human pose fitting by calibrating body shape once, then using a shape-conditioned point diffusion prior, PointDiT, to guide optimization, improving absolute and local pose accuracy on EMDB.

  2. PFM-HR: Pose Flow Matching for Humanoid Robots

    cs.RO 2026-08 conditional novelty 6.0 of 10

    A flow-matching pose prior, kept frozen, provides a Jacobian-based directional score that modulates tracking rewards and improves RL-based humanoid motion tracking, especially for dynamic acrobatic skills.

  3. Geometric Neural Distance Fields for Learning Human Motion Priors

    cs.CV 2025-09 conditional novelty 6.0 of 10

    NRMF is a human motion prior that represents plausible pose, velocity, and acceleration as zero-level sets of neural distance fields and improves motion denoising, in-betweening, and fitting to 2D/3D observations.

  4. SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidance

    cs.CV 2026-04 unverdicted novelty 5.0 of 10

    SocialMirror reconstructs 3D meshes of closely interacting humans from monocular videos using semantic guidance from vision-language models and geometric constraints in a diffusion model to handle occlusions and maint...

  5. FashionPose: Text to Pose to Relight Image Generation for Personalized Fashion Visualization

    cs.CV 2025-07 reject novelty 4.0 of 10

    A single caption can drive pose generation, person-image synthesis, and relighting through a three-stage FashionPose pipeline, with reported text-to-pose gains on DF-PASS that are undermined by inconsistent tables.

Pith tools