nuScenes

Introduced by Caesar et al. in nuScenes: A multimodal dataset for autonomous driving

The nuScenes dataset is a large-scale autonomous driving dataset. The dataset has 3D bounding boxes for 1000 scenes collected in Boston and Singapore. Each scene is 20 seconds long and annotated at 2Hz. This results in a total of 28130 samples for training, 6019 samples for validation and 6008 samples for testing. The dataset has the full autonomous vehicle data suite: 32-beam LiDAR, 6 cameras and radars with complete 360° coverage. The 3D object detection challenge evaluates the performance on 10 classes: cars, trucks, buses, trailers, construction vehicles, pedestrians, motorcycles, bicycles, traffic cones and barriers.

Source: PointPainting: Sequential Fusion for 3D Object Detection

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
3D Object Detection	nuScenes	EA-LSS
3D Multi-Object Tracking	nuScenes	Fast-Poly
Trajectory Prediction	nuScenes	UniTraj
LIDAR Semantic Segmentation	nuScenes	PTv3 + PPT
3D Object Detection	nuScenes Camera Only	Far3D
Bird's-Eye View Semantic Segmentation	nuScenes	PointBeV
Semi-Supervised Semantic Segmentation	nuScenes	LaserMix
3D Object Detection	nuscenes Camera-Radar	HyDRa
Lane Detection	nuScenes	DSLP
3D Semantic Segmentation	nuScenes	PTv2
3D Multi-Object Tracking	nuscenes Camera-Radar	HyDRa
Object Detection	nuScenes	RANet
Motion Detection	nuScenes	FastFlowNet
HD semantic map learning	nuScenes	VectorMapNet
3D Multi-Object Tracking	nuScenes Camera Only	StreamPETR-Large
Trajectory Planning	nuScenes	AD-MLP
Motion Planning	nuScenes	GPT-Driver
Prediction Of Occupancy Grid Maps	nuScenes	TPVFormer04
Weakly supervised Semantic Segmentation	nuScenes	COARSE3D
Instance Segmentation	nuScenes	TraDeS