mlsanityData to 14 Oct 2026

PEARL: A Lightweight Prompt-based Feature Interpreter Framework for Real-Time, Anonymous, and Heterogeneous Collaborative Perception

arXiv 2609.27123 ↗First posted 22 Sep 2026cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Heterogeneity across Collaborative Perception (CP) agents is a major challenge for emerging CP frameworks due to domain gaps from differing sensors, architectures, and training data. Prior works mitigate this challenge by aligning features in a unified space via model retraining or per-agent-type interpreters. These strategies (a) require access to neighbor configurations, (b) do not fully address real-time CP deployment, and (c) generalize poorly to unseen agents joining at run time. To overcome these challenges, we present PEARL, a Prompt-Embedding framework for Anonymous and Real-time Lightweight heterogeneous CP. PEARL supports multiple CP interpreters and selects one for a new-joining agent in real time using two lightweight, multi-scale interpreters trained in parallel: a sparse-detection (LWSD) interpreter that aligns salient regions for cooperative detection, and a dense, domain-invariant (LWDDI) interpreter that produces agent-invariant features for fast interpreter selection. Both interpreters use low-rank visual prompts to reduce computation, storage, and model complexity. Extensive experiments on simulated (OPV2V, V2XSet) and real (DAIR-V2X) datasets show that PEARL generalizes across simulated and real-world cooperative driving scenarios. Its real-time model-selection strategy yields an 8.2% Average Precision (AP) gain over a random-selection baseline while running in 1.67 ms on average. Although primarily designed for real-time CP, PEARL also outperforms state-of-the-art heterogeneous CP frameworks under traditional offline training by 5.6% AP on average while reducing communication cost by up to 34.7 times. Equally important, PEARL does not require sharing agents' configurations or model settings, thereby protecting information that may be proprietary or private. These results establish PEARL as a scalable and practical framework for heterogeneous collaborative perception.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 30 of its 42 references that are in the map. First posted on arXiv 22 Sep 2026.

Builds on

30 of 42 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  3. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,715
  4. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  5. 2022V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformercited by 320
  6. 2021OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communicationcited by 304
  7. 2022DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detectioncited by 273
  8. 2022Where2comm: Communication-Efficient Collaborative Perception via Spatial Confidence Mapscited by 255
  9. 2022CoBEVT: Cooperative Bird's Eye View Semantic Segmentation with Sparse Transformerscited by 227
  10. 2022Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptationcited by 98
  11. 2024An Extensible Framework for Open Heterogeneous Collaborative Perceptioncited by 87
  12. 2023HM-ViT: Hetero-modal Vehicle-to-Vehicle Cooperative perception with vision transformercited by 70
18 more
  1. 2024Communication-Efficient Collaborative Perception via Information Filling with Codebookcited by 68
  2. 2023AD-CLIP: Adapting Domains in Prompt Space Using CLIPcited by 62
  3. 2022Bridging the Domain Gap for Multi-Agent Perceptioncited by 60
  4. 2023DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detectioncited by 34
  5. 2023Exploring Sparse Visual Prompt for Domain Adaptive Dense Predictioncited by 28
  6. 2023Domain-Controlled Prompt Learningcited by 22
  7. 2023Learning Domain-Aware Detection Head with Prompt Tuningcited by 18
  8. 2024One is Plenty: A Polymorphic Feature Interpreter for Immutable Heterogeneous Collaborative Perceptioncited by 15
  9. 2023A Spatial Calibration Method for Robust Cooperative Perceptioncited by 13
  10. 2024A Survey on Intermediate Fusion Methods for Collaborative Perception Categorized by Real World Challengescited by 13
  11. 2024Robust Collaborative Perception without External Localization and Clock Devicescited by 8
  12. 2024DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detectioncited by 8
  13. 2025NegoCollab: A Common Representation Negotiation Approach for Heterogeneous Collaborative Perceptioncited by 5
  14. 2025UPRE: Zero-Shot Domain Adaptation for Object Detection via Unified Prompt and Representation Enhancementcited by 4
  15. 2025CoBEVMoE: Heterogeneity-aware Feature Fusion with Dynamic Mixture-of-Experts for Collaborative Perceptioncited by 3
  16. 2024Self-Localized Collaborative Perceptioncited by 2
  17. 2026Faster-HEAL: An Efficient and Privacy-Preserving Collaborative Perception Framework for Heterogeneous Autonomous Vehiclescited by 2
  18. 2025Prmpt2Adpt: Prompt-Based Zero-Shot Domain Adaptation for Resource-Constrained Environmentscited by 1

Built on by

Not cited yet.