52 datasets found

KITTI 3D object detection dataset, containing 7481 training images and 7518 testing images.

Stereo CenterNet is a 3D object detection method for stereo images, combining deep learning and geometry.

The Waymo Open Dataset and the Waymo Open Motion Dataset are used in this paper.

The KITTI Object Detection Benchmark consists of 7,481 training images and 7,518 testing images, with 3D LiDAR point clouds and camera images.

The REAL dataset is a real-world dataset capturing 4300 real-world images of 7 scenes for training, and 2750 real-world images of 6 scenes for evaluation.

The CAMERA dataset is a synthetic dataset generated by rendering and compositing objects into real scenes in a context-aware manner.

Waymo Open Dataset (validation set)

FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformer

LiDAR-based 3D object detection dataset

A benchmark for stereo matching and depth estimation.

DiffRef3D:ADiffusion-basedProposalRefinementFrameworkfor3DObjectDetection

Monocular 3D object localization in driving scenes is a crucial task, but challenging due to its ill-posed nature. Estimating 3D coordinates for each pixel on the object surface...

KITTI 2012 is a real-world dataset in the outdoor scenario, and contains 194 training and 195 testing stereo image pairs with the size of 376 × 1240.

A LiDAR-based 3D object detection dataset.

A LiDAR-based 3D object detection dataset.

Multi-agent collaborative perception dataset and benchmark for autonomous driving

Vehicle-infrastructure cooperative 3D object detection dataset

Aerial-ground collaborative perception framework for 3D object detection

The KITTI Vision Benchmark Suite is a dataset used for object detection and tracking in autonomous vehicles.

The Waymo Open dataset provides 798 training sequences and 202 validation sequences. Each sequence contains 200 frames within 20 seconds.