Abstract:Modern detectors are mostly trained under single and limited conditions. However, object detection faces various complex and open situations in autonomous driving, especially in urban street scenes with dense objects and complex backgrounds. Due to the shift in data distribution, modern detectors can not perform well in actual urban environments. Using domain adaptation to improve detection performance is one of the key methods to extend object detection from limited situations to open situations. To this end, this article proposes a Domain Adaptation of Anchor-Free object detection (DAAF) for urban traffic. DAAF is a cross-domain object detection method that performs feature alignment including two aspects. On the one hand, we designed a fully convolutional adversarial training method for global feature alignment at the image level. Meanwhile, images can generally be decomposed into structural information and texture information. In urban street scenes, the structural information of images is generally similar. The main difference between the source domain and the target domain is texture information. Therefore, during global feature alignment, this paper proposes a method called texture information limitation (TIL). On the other hand, in order to solve the problem of variable aspect ratios of objects in urban street scenes, this article uses an anchor-free detector as the baseline detector. Since the anchor-free object detector can obtain neither explicit nor implicit instance-level features, we adopt Pixel-Level Adaptation (PLA) to align local features instead of instance-level alignment for local features. The size of the object has the greatest impact on the final detection effect, and the object scale in urban scenes is relatively rich. Guided by the differentiation of attention mechanisms, a multi-level adversarial network is designed to perform feature alignment of the output space at different feature levels called Scale Information Limitation (SIL). We conducted cross-domain detection experiments by using various urban streetscape autonomous driving object detection datasets, including adverse weather conditions, synthetic data to real data, and cross-camera adaptation. The experimental results indicate that the method proposed in this article is effective.

Domain Adaptive Object Detection for Autonomous Driving under Foggy Weather

Domain Adaptation based Object Detection for Autonomous Driving in Foggy and Rainy Weather

Joint Feature-Level And Pixel-Level Domain Adaption For Object Detection In The Wild

Progressive Domain Adaptive Object Detection Based on Self‐Attention in Foggy Weather

Object Detection in Foggy Scenes by Embedding Depth and Reconstruction into Domain Adaptation

Object Detection in Adverse Weather for Autonomous Vehicles Based on Sensor Fusion and Incremental Learning

DA-RAW: Domain Adaptive Object Detection for Real-World Adverse Weather Conditions

Multi-Target Domain Adaptation via Unsupervised Domain Classification for Weather Invariant Object Detection

YOLO-Vehicle-Pro: A Cloud-Edge Collaborative Framework for Object Detection in Autonomous Driving under Adverse Weather Conditions

Prior-Based Domain Adaptive Object Detection for Hazy and Rainy Conditions

Domain adaptation of anchor-free object detection for urban traffic

DAGL-Faster: Domain adaptive faster r-cnn for vehicle object detection in rainy and foggy weather conditions

Brain-Inspired Domain-Incremental Adaptive Detection for Autonomous Driving

Adversarial Defense Teacher for Cross-Domain Object Detection under Poor Visibility Conditions

A Step-Wise Domain Adaptation Detection Transformer for Object Detection under Poor Visibility Conditions

YOLOv5s-Fog: An Improved Model Based on YOLOv5s for Object Detection in Foggy Weather Scenarios

AIE-YOLO: Effective object detection method in extreme driving scenarios via adaptive image enhancement

C2FDA: Coarse-to-Fine Domain Adaptation for Traffic Object Detection

Enhancing the Safety of Autonomous Vehicles in Adverse Weather by Deep Learning-Based Object Detection

Multi-Level Domain Adaptive Learning For Cross-Domain Detection

Multiple Adverse Weather Conditions Adaptation for Object Detection via Causal Intervention