SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception

2026-09-28

AI Summary

This study addresses the challenges of long-range feature redundancy and communication inefficiency in V2X cooperative perception by proposing a compact evidence transmission framework based on closed-form fidelity. Methodologically, the rate-distortion function is decomposed into an irreducible core to establish theoretical zero-distortion bounds, guiding the design of precise BEV support selection, pointwise channel compression, and masked weight-sharing reconstruction mechanisms for efficient context recovery under bounded inference. Evaluated on the V2XSet benchmark, the proposed approach reduces feature payload by 26.6× while improving Average Precision by 4.13 and 8.57 percentage points, with only 3.81% additional latency. These results demonstrate a significant breakthrough in overcoming communication efficiency bottlenecks for collaborative autonomous driving systems.

Abstract

Vehicle-to-Everything (V2X) cooperative perception improves 3-D detection by sharing intermediate features, but dense remote features may repeat context that the ego agent can infer locally. Most communication-efficient designs optimize masks or codes empirically, leaving a more basic question open: which remote evidence is indispensable given the receiver's own observation? We introduce a closure-fidelity perspective on ego conditioned remote perception. Under a finite deductive abstraction and explicit conditions, its rate--distortion function decomposes over an irredundant core, and the exact zero-distortion rate becomes $P_A H(π_A)$. This analysis suggests a concrete design principle: transmit compact evidence and recover derivable context with bounded receiver-side inference. Guided by this principle, SemRD-V2X is an operational neural proxy that combines exact-budget BEV support selection, pointwise channel compression, and masked shared-weight reconstruction before standard fusion. Experiments on simulated V2XSet and real-world DAIR-V2X validate the resulting design. In a controlled five-run V2XSet comparison against a locally reproduced V2X-ViT-v1 baseline on one Tesla V100, SemRD-V2X reduces the analytical feature payload by $26.6\times$ while improving AP@0.5/AP@0.7 by 4.13/8.57 points, with 3.81\% additional mean compute latency. These results position closure fidelity as both an analytical lens and an actionable design principle for communication-efficient cooperative perception.

Problem

• Cooperative Perception
• V2X Communication
• Redundancy Elimination
• Communication Efficiency
• Rate-Distortion

Innovation

• Closure-Fidelity
• Rate-Distortion
• Cooperative Perception
• V2X Communication
• Bounded Inference