computer-vision-10

Bird's-Eye View Semantic Segmentation

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in birds-eye-view-semantic-segmentation-21

Trend

Dataset

Best Model

Actions

nuScenes

Lyft Level 5

Libraries

i

Use these libraries to find birds-eye-view-semantic-segmentation-21 models and implementations

Datasets

nuScenes

LandCover.ai

Subtasks

No subtasks available.

Most implemented papers

BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers

Wenhai Wang, Enze Xie, Tong Lu, Jifeng Dai, Chonghao Sima, Hongyang Li, Zhiqi Li, Qiao Yu•Wed Mar 30 2022

This work presents a new framework termed BEVFormer, which learns unified BEV representations with spatiotemporal transformers to support multiple autonomous driving perception tasks and remarkably improves the accuracy of velocity estimation and recall of objects under low visibility conditions.

1710

Content

0

Paper Graph

Cross-view Transformers for real-time Map-view Semantic Segmentation

Brady Zhou, Philipp Krahenbuhl•Wed May 04 2022

The architecture implicitly learns a mapping from individual camera views into a canonical map-view representation using a camera-aware cross-view attention mechanism, performing at state-of-the-art on the nuScenes dataset, with 4x faster inference speeds.

347 0

Paper Graph

CoBEVT: Cooperative Bird's Eye View Semantic Segmentation with Sparse Transformers

Bolei Zhou, Zhengzhong Tu, Runsheng Xu, Jiaqi Ma, Hao Xiang, Wei Shao•Mon Jul 04 2022

The extensive experiments on the V2V perception dataset, OPV2V, demonstrate that CoBEVT achieves state-of-the-art performance for cooperative BEV semantic segmentation and is shown to be generalizable to other tasks, including 1) BEV segmentation with single-agent multi-camera and 2) 3D object detection with multi-agent LiDAR systems.

318 0

Paper Graph

MatrixVT: Efficient Multi-Camera to BEV Transformation for 3D Perception

Xiangyu Zhang, Zeming Li, Zheng Ge, Hongyu Zhou•Fri Nov 18 2022

This paper proposes an efficient multi-camera to Bird’s-Eye-View (BEV) view transformation method for 3D perception, dubbed MatrixVT, which enjoys a faster speed and less memory footprint while remaining deploy-friendly.

58 0

Paper Graph

LandCover.ai: Dataset for Automatic Mapping of Buildings, Woodlands, Water and Roads from Aerial Imagery

Adrian Boguszewski, D. Batorski, Natalia Ziemba-Jankowska, T. Dziedzic, Anna Zambrzycka•Mon May 04 2020

The introduction of LandCover.ai (Land Cover from Aerial Imagery) dataset for semantic segmentation proves that the automatic mapping of land cover is possible with a relatively small, cost-efficient, RGB-only dataset.

114 0

Paper Graph

Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D

S. Fidler, Jonah Philion•Wed Aug 12 2020

In pursuit of the goal of learning dense representations for motion planning, it is shown that the representations inferred by the model enable interpretable end-to-end motion planning by "shooting" template trajectories into a bird's-eye-view cost map output by the network.

1408 0

Paper Graph

PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images

X. Zhang, Junjie Yan, Jian Sun, Yingfei Liu, Fan Jia, Shuailin Li, Qi Gao, Tiancai Wang•Wed Jun 01 2022

PETRv2, a unified framework for 3D perception from multi-view images, is proposed and the 3D position embedding in PETR is extended for temporal modeling, which achieves the temporal alignment on object position of different frames.

474 0

Paper Graph

Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?

Rares Ambrus, Adam W. Harley, Zhaoyuan Fang, Katerina Fragkiadaki, Jie Li•Wed Jun 15 2022

It is found that batch size and input resolution greatly affect performance, while lifting strategies have a more modest effect, and it is demonstrated that radar data can provide a substantial boost to performance, helping to close the gap between camera-only and LiDAR-enabled systems.

175 0

Paper Graph

LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentation

M. Cord, Alahari Karteek, Florent Bartoccioni, 'Eloi Zablocki, Andrei Bursuc, Patrick P'erez•Sun Jun 26 2022

'LaRa' is presented, an efficient encoder-decoder, transformer-based model for vehicle semantic segmentation from multiple cameras that outperforms the best previous works using transformers on nuScenes.

40 0

Paper Graph

Adding a benchmark result helps the community track progress.