ML Research Wiki / Benchmarks / Bird's-Eye View Semantic Segmentation / nuScenes

nuScenes

Bird's-Eye View Semantic Segmentation Benchmark

Performance Over Time

📊 Showing 15 results | 📏 Metric: IoU veh - 224x480 - Vis filter. - 100x100 at 0.5

Top Performing Models

Rank Model Paper IoU veh - 224x480 - Vis filter. - 100x100 at 0.5 Date Code
1 PETRv2 PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images 44.80 2022-06-02 📦 megvii-research/petr
2 MatrixVT MatrixVT: Efficient Multi-Camera to BEV Transformation for 3D Perception 44.80 2022-11-19 📦 megvii-basedetection/bevdepth 📦 ZRandomize/MatrixVT
3 PointBeV PointBeV: A Sparse Approach to BeV Predictions 39.90 2023-12-01 📦 valeoai/pointbev
4 PointBeV (static) PointBeV: A Sparse Approach to BeV Predictions 38.70 2023-12-01 📦 valeoai/pointbev
5 FIERY FIERY: Future Instance Prediction in Bird's-Eye View from Surround Monocular Cameras 38.20 2021-04-21 📦 wayveai/fiery
6 Simple-BEV Simple-BEV: What Really Matters for Multi-Sensor BEV Perception? 36.90 2022-06-16 📦 valeoai/pointbev
7 BEVFormer BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers 35.80 2022-03-31 📦 fundamentalvision/BEVFormer 📦 valeoai/pointbev 📦 zhiqi-li/BEVFormer
8 FIERY (static) FIERY: Future Instance Prediction in Bird's-Eye View from Surround Monocular Cameras 35.80 2021-04-21 📦 wayveai/fiery
9 LaRa LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentation 35.40 2022-06-27 📦 valeoai/LaRa
10 CVT Cross-view Transformers for real-time Map-view Semantic Segmentation 31.40 2022-05-05 📦 bradyz/cross_view_transformers 📦 valeoai/pointbev

All Papers (15)

megvii-research/petr
megvii-basedetection/bevdepth ZRandomize/MatrixVT
valeoai/pointbev
valeoai/pointbev
wayveai/fiery
valeoai/pointbev
fundamentalvision/BEVFormer valeoai/pointbev zhiqi-li/BEVFormer
wayveai/fiery
valeoai/LaRa
bradyz/cross_view_transformers valeoai/pointbev
nv-tlabs/lift-splat-shoot
opendrivelab/st-p3