158 datasets found

The Common Pets in 3D (COP3D) dataset contains casual, in-the-wild videos of pets, and used it to learn priors for novel view synthesis in dynamic scenes.

The ShapeNet-ViPC dataset contains 38,328 objects from 13 categories; for each object it comprises 24 partial point clouds with occlusions generated under 24 viewpoints.

A high-quality dataset containing 348 different car models with physical dimensions and part-level annotations based on global and local deformations.

HoliCity is a city-scale 3D dataset with rich structural information. It contains 6,300 real-world panoramas of resolution 13312 × 6656 that are accurately aligned with the CAD...

The dataset used in the paper is a 2D and 3D urban environment.

MonoSelfRecon: Purely Self-Supervised Explicit Generalizable 3D Reconstruction of Indoor Scenes from Monocular RGB Views

Surface Light Field Fusion dataset for interactive scanning of highly reflective objects with a commodity RGBD sensor.

The Waymo Open dataset is a large-scale video dataset used for testing object-centric video models.

Sparse Inertial Poser: Automatic 3D Human Pose Estimation from Sparse IMUs

A dataset for multi-view stereo reconstruction.

A dataset for 3D object reconstruction from a single image.

A new method for 3D shape reconstruction from a single image, in which a deep neural network directly maps an image to a vector of network weights.

Real and synthesized mmWave radar data, ground truth 2D depth images and point clouds for 3DRIMR experiments

Meshroom is a 3D reconstruction software.

SDFStudio is a unified framework for surface reconstruction.

The ETH3D dataset is a large-scale dataset for multi-view stereo reconstruction. It contains 120 scenes with complex geometric layouts.

The Middlebury and ETH3D datasets are used for stereo matching and 3D reconstruction tasks.

Neural implicit representations have gained popularity in multi-view 3D reconstruction. However, most previous work struggles to yield physically plausible results, limiting...

The Mip-NeRF 360 and Tanks & Temples datasets are used to evaluate the performance of the Pixel-GS method.

The dataset used in this paper is a set of images of a 1D target acquired by a stereo vision system with different distortion models.