AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer

Runsheng Xu, Hao Xiang, Zhengzhong Tu, Xin Xia, Ming-Hsuan Yang, Jiaqi Ma

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 671 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add multi-scale window attention (MSwin) (ablation)V2X-ViT base · V2XSet NoisyAP@0.7: 0.478 -> 0.519Improved
add split attention fusion of window branches (ablation)V2X-ViT base + MSwin · V2XSet NoisyAP@0.7: 0.519 -> 0.548Improved
add heterogeneous multi-agent self-attention (HMSA) (ablation)V2X-ViT base + MSwin + SpAttn · V2XSet NoisyAP@0.7: 0.548 -> 0.601Improved
add delay-aware positional encoding (DPE) (ablation)V2X-ViT · V2XSet NoisyAP@0.7: 0.601 -> 0.614Improved
add DPE under 100 ms delay (ablation)V2X-ViT · V2XSet Noisy (100 ms delay)AP@0.7: 0.639 -> 0.650Improved
add DPE under 400 ms delay (ablation)V2X-ViT · V2XSet Noisy (400 ms delay)AP@0.7: 0.458 -> 0.478Improved
use 3 encoder layers instead of 1 (ablation)V2X-ViT · V2XSet NoisyAP@0.7: 0.591 -> 0.614Improved
transform features to ego range before cropping instead of broadcasting (ablation)V2X-ViT · V2XSet PerfectAP@0.7: 0.623 -> 0.712Improved
transform features to ego range before cropping instead of broadcasting (ablation)DiscoNet · V2XSet PerfectAP@0.7: 0.610 -> 0.695Improved
add noise (pose error and 100 ms delay) to evaluationEarly Fusion · V2XSetAP@0.7: 0.710 -> 0.384Hurt
add noise (pose error and 100 ms delay) to evaluationLate Fusion · V2XSetAP@0.7: 0.620 -> 0.307Hurt
add noise (pose error and 100 ms delay) to evaluationV2X-ViT · V2XSetAP@0.7: 0.712 -> 0.614Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy