Saved in:
Bibliographic Details
Main Authors: Kawaguchi, Joshua, Manzur, Saad, Wang, Emily Gao, Sinha, Maitreyi, Vela, Bryan, Wang, Yunxi, Vela, Brandon, Hayes, Wayne B.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.00991
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911350995812352
author Kawaguchi, Joshua
Manzur, Saad
Wang, Emily Gao
Sinha, Maitreyi
Vela, Bryan
Wang, Yunxi
Vela, Brandon
Hayes, Wayne B.
author_facet Kawaguchi, Joshua
Manzur, Saad
Wang, Emily Gao
Sinha, Maitreyi
Vela, Bryan
Wang, Yunxi
Vela, Brandon
Hayes, Wayne B.
contents Diverse, accurately labeled 3D human pose data is expensive and studio-bound, while in-the-wild datasets lack known ground truth. We introduce UnrealPose-Gen, an Unreal Engine 5 pipeline built on Movie Render Queue for high-quality offline rendering. Our generated frames include: (i) 3D joints in world and camera coordinates, (ii) 2D projections and COCO-style keypoints with occlusion and joint-visibility flags, (iii) person bounding boxes, and (iv) camera intrinsics and extrinsics. We use UnrealPose-Gen to present UnrealPose-1M, an approximately one million frame corpus comprising eight sequences: five scripted "coherent" sequences spanning five scenes, approximately 40 actions, and five subjects; and three randomized sequences across three scenes, approximately 100 actions, and five subjects, all captured from diverse camera trajectories for broad viewpoint coverage. As a fidelity check, we report real-to-synthetic results on four tasks: image-to-3D pose, 2D keypoint detection, 2D-to-3D lifting, and person detection/segmentation. Though time and resources constrain us from an unlimited dataset, we release the UnrealPose-1M dataset, as well as the UnrealPose-Gen pipeline to support third-party generation of human pose data.
format Preprint
id arxiv_https___arxiv_org_abs_2601_00991
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle UnrealPose: Leveraging Game Engine Kinematics for Large-Scale Synthetic Human Pose Data
Kawaguchi, Joshua
Manzur, Saad
Wang, Emily Gao
Sinha, Maitreyi
Vela, Bryan
Wang, Yunxi
Vela, Brandon
Hayes, Wayne B.
Computer Vision and Pattern Recognition
Diverse, accurately labeled 3D human pose data is expensive and studio-bound, while in-the-wild datasets lack known ground truth. We introduce UnrealPose-Gen, an Unreal Engine 5 pipeline built on Movie Render Queue for high-quality offline rendering. Our generated frames include: (i) 3D joints in world and camera coordinates, (ii) 2D projections and COCO-style keypoints with occlusion and joint-visibility flags, (iii) person bounding boxes, and (iv) camera intrinsics and extrinsics. We use UnrealPose-Gen to present UnrealPose-1M, an approximately one million frame corpus comprising eight sequences: five scripted "coherent" sequences spanning five scenes, approximately 40 actions, and five subjects; and three randomized sequences across three scenes, approximately 100 actions, and five subjects, all captured from diverse camera trajectories for broad viewpoint coverage. As a fidelity check, we report real-to-synthetic results on four tasks: image-to-3D pose, 2D keypoint detection, 2D-to-3D lifting, and person detection/segmentation. Though time and resources constrain us from an unlimited dataset, we release the UnrealPose-1M dataset, as well as the UnrealPose-Gen pipeline to support third-party generation of human pose data.
title UnrealPose: Leveraging Game Engine Kinematics for Large-Scale Synthetic Human Pose Data
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.00991