paper-with-me

홈 › Papers

To The Point: Correspondence-driven monocular 3D category reconstruction

2021-06-10 · NeurIPS 2021 12 · Filippos Kokkinos, Iasonas Kokkinos

We present To The Point (TTP), a method for reconstructing 3D objects from a single image using 2D to 3D correspondences learned from weak supervision. We recover a 3D shape from a 2D image by first regressing the 2D positions corresponding to the 3D template vertices and then jointly estimating a rigid camera transform and non-rigid template deformation that optimally explain the 2D positions through the 3D shape projection. By relying on 3D-2D correspondences we use a simple per-sample optimization problem to replace CNN-based regression of camera pose and non-rigid deformation and thereby obtain substantially more accurate 3D reconstructions. We treat this optimization as a differentiable layer and train the whole system in an end-to-end manner. We report systematic quantitative improvements on multiple categories and provide qualitative results comprising diverse shape, pose and texture prediction examples. Project website: https://fkokkinos.github.io/to_the_point/.

📄 PDF Abstract BibTeX arXiv:2106.05662

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

ViSER: Video-Specific Surface Embeddings for Articulated 3D Shape Reconstruction

2021-12-01 · NeurIPS 2021 12 · Gengshan Yang, Deqing Sun, Varun Jampani, Daniel Vlasic 외

We introduce ViSER, a method for recovering articulated 3D shapes and dense3D trajectories from monocular videos. Previous work on high-quality reconstruction of dynamic 3D shapes typically relies on multiple camera vie…

3D Shape Reconstruction from Videos

fCOP: Focal Length Estimation from Category-level Object Priors

2024-09-29 · Xinyue Zhang, Jiaqi Yang, Xiangting Meng, Abdelrahman Mohamed 외

In the realm of computer vision, the perception and reconstruction of the 3D world through vision signals heavily rely on camera intrinsic parameters, which have long been a subject of intense research within the communi…

Depth EstimationMonocular Depth EstimationObjectRepresentation Learning

Category-Level 3D Correspondence in Camera Space via Morphable Object Priors

2026-05-27 · Leonhard Sommer, Artur Jesslen, Basavaraj Sunagad, Adam Kortylewski arxiv

Understanding 3D objects from images is fundamental to robotics and AR/VR applications. While recent work has made progress in category-level pose estimation, current representations fail to capture the fine-grained sema…

Pose Estimation

DynOMo: Online Point Tracking by Dynamic Online Monocular Gaussian Reconstruction

2024-09-03 · Jenny Seidenschwarz, Qunjie Zhou, Bardienus Duisterhof, Deva Ramanan 외

Reconstructing scenes and tracking motion are two sides of the same coin. Tracking points allow for geometric reconstruction [14], while geometric reconstruction of (dynamic) scenes allows for 3D tracking of points over …

Mixed RealityMonocular ReconstructionPoint TrackingRobot Navigation

LongDPM: Overlap-Aware 4D Reconstruction from Long Monocular Videos

2026-05-17 · Chenyi Xu, Yihao Wu, Liqi Yan, Chao Yang 외 arxiv

Recovering a dynamic 3D scene from a long monocular video is crucial for dense geometry, camera motion, and temporal correspondence to remain consistent in a shared coordinate system. Existing methods face two key challe…

Dynamic ReconstructionCamera Pose Estimation