paper-with-me

Papers

NeuPhysics: Editable Neural Geometry and Physics from Monocular Videos

2022-10-22 · Yi-Ling Qiao, Alexander Gao, Ming C. Lin

We present a method for learning 3D geometry and physics parameters of a dynamic scene from only a monocular RGB video input. To decouple the learning of underlying scene geometry from dynamic motion, we represent the scene as a time-invariant signed distance function (SDF) which serves as a reference frame, along with a time-conditioned deformation field. We further bridge this neural geometry representation with a differentiable physics simulator by designing a two-way conversion between the neural field and its corresponding hexahedral mesh, enabling us to estimate physics parameters from the source video by minimizing a cycle consistency loss. Our method also allows a user to interactively edit 3D objects from the source video by modifying the recovered hexahedral mesh, and propagating the operation back to the neural field representation. Experiments show that our method achieves superior mesh and video reconstruction of dynamic scenes compared to competing Neural Field approaches, and we provide extensive examples which demonstrate its ability to extract useful 3D representations from videos captured with consumer-grade cameras.

📄 PDF Abstract BibTeX arXiv:2210.12352

Code (0)

등록된 구현이 없습니다.

Tasks

3D geometryVideo Reconstruction

Similar Papers 제목 키워드 기반

MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos

2026-05-28 · Daniel Rho, Jun Myeong Choi, Matthew Thornton, Biswadip Dey 외 arxiv

Existing inverse physics methods recover physical parameters from multi-view videos, where geometric constraints across views resolve scale and 3D structure. In monocular settings, however, such constraints are absent, l…

Creating Your Editable 3D Photorealistic Avatar with Tetrahedron-constrained Gaussian Splatting

2025-04-29 · CVPR 2025 1 · Hanxi Liu, Yifang Men, Zhouhui Lian

Personalized 3D avatar editing holds significant promise due to its user-friendliness and availability to applications such as AR/VR and virtual try-ons. Previous studies have explored the feasibility of 3D editing, but …

Representation Learning

CRISP: Contact-Guided Real2Sim from Monocular Video with Planar Scene Primitives

2025-12-16 · Zihan Wang, Jiashun Wang, Jeff Tan, Yiwen Zhao 외 arxiv

We introduce CRISP, a method that recovers simulatable human motion and scene geometry from monocular video. Prior work on joint human-scene reconstruction relies on data-driven priors and joint optimization with no phys…

Reinforcement Learning

AeroDGS: Physically Consistent Dynamic Gaussian Splatting for Single-Sequence Aerial 4D Reconstruction

2026-02-25 · Hanyang Liu, Rongjun Qin arxiv

Recent advances in 4D scene reconstruction have significantly improved dynamic modeling across various domains. However, existing approaches remain limited under aerial conditions with single-view capture, wide spatial r…

Physics-based Human Pose Estimation from a Single Moving RGB Camera

2025-07-23 · Ayce Idil Aytekin, Chuqiao Li, Diogo Luvizon, Rishabh Dabral 외 arxiv

Most monocular and physics-based human pose tracking methods, while achieving state-of-the-art results, suffer from artifacts when the scene does not have a strictly flat ground plane or when the camera is moving. Moreov…

Pose EstimationPose Tracking