V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction

By Tsun-Hsuan Wang, Sivabalan Manivasagam, Ming Liang, Binh Yang, Wenyuan Zeng, James Tu, Raquel Urtasun. European Conference on Computer Vision, 2020

654 citations recorded by the source corpus.

In this paper, we explore the use of vehicle-to-vehicle (V2V) communication to improve the perception and motion forecasting performance of self-driving vehicles. By intelligently aggregating the information received from multiple nearby vehicles, we can observe the same scene from different viewpoints. This allows us to see through occlusions and detect actors at long range, where the observations are very sparse or non-existent. We also show that our approach of sending compressed deep feature map activations achieves high accuracy while satisfying communication bandwidth requirements.

Subfields: Automotive Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/728f3e5932cc5c7bc617c72d2bcc7aad658dfab0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Where2comm: Communication-Efficient Collaborative Perception via Spatial Confidence Maps (2022, methodology): “For example, When2com [12] considers a handshake mechanism which selects the most relevant collaborators; V2VNet [1] considers end-to-end-learning-based source coding; and DiscoNet [2] uses 1D convolution to compress message.”
CoBEVT: Cooperative Bird's Eye View Semantic Segmentation with Sparse Transformers (2022, methodology): “We compare with the state-of-the-art multi-agent perception algorithms: F-Cooper [21], AttFuse [12], V2VNet [20], and DiscoNet [14].”
Learning Distilled Collaboration Graph for Multi-Agent Perception (2021, background): “DiscoNet(64) achieves 192 times less communication volume and still outperforms V2VNet [34].”
VoxFormer: Sparse Voxel Transformer for Camera-Based 3D Semantic Scene Completion (2023, background): “Holistic 3D scene understanding is challenged by limited sensing range, and researchers have proposed multi-agent collaborative perception to introduce more observations of the 3D scene [43– 49].”
V2V4Real: A Real-World Large-Scale Dataset for Vehicle-to-Vehicle Cooperative Perception (2023, methodology): “For instance, V2VNet [40] adopted graph neural networks to fuse intermediate features.”
DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection (2022, background): “V2VNet [25] is a pioneering work in V2V that broadcasts compressed intermediate features and propagates message received from nearby vehicles to generate motion forecasts.”
OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication (2021, methodology): “Unlike the single vehicle’s perception area where multiple large-scale public datasets exist [12], [13], [14], most of the current V2V perception algorithms conduct experiments based on their customized data [15], [16], [17].”