Domain Adaptive Object Detection for Autonomous Driving under Foggy Weather

Li, Jinlong; Xu, Runsheng; Ma, Jin; Zou, Qin; Yu, Hongkai

doi:10.1109/wacv56688.2023.00068

Cited by 64 publications

(20 citation statements)

References 81 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clearly, Due to all methods unseen these data with V2V implementation gap in the training phase, their detection performance is even worse than single-agent perception baseline NO Fusion on for AP@0. [28] and AdvGRL [14], all methods have the improvement of performance. For example, V2X-ViT improved by 12.1%/5.3% for AP@0.5/0.7 with GRL, and 12.7%/5.8% for AP@0.5/0.7 with AdvGRL.…”

Section: Quantitaive Evaluationmentioning

confidence: 90%

“…To demonstrate the significant effect of implementation gap and domain gap, we first train these methods on perfect setting of OPV2V training set, Then, these methods are evaluated on Noisy Setting of OPV2V testing set and V2V4Real testing set to assess their performance. In addition, to show the effectiveness of our proposed domain adaptation modules on Feature-Gap scenario, two domain adaptation methods i.e., gradient reverse layer (GRL) [28] and adversarial gradient reverse layer (AdvGRL) [14] are utilized to backpropagate the gradient to assist the model for generating domain-invariant features by two domain classifiers including feature-level and objectlevel classifiers.…”

Section: B Experiments Setupmentioning

confidence: 99%

See 1 more Smart Citation

Decision Support Multi-agent Modeling and Simulation of Aeronautic Marine Oil Spill Response

Liu

Tian

et al. 2023

Smart Innovation, Systems and Technologies

View full text Add to dashboard Cite

Due to the lack of real multi-agent data and timeconsuming of labeling, existing multi-agent cooperative perception algorithms usually select the simulated sensor data for training and validating. However, the perception performance is degraded when these simulation-trained models are deployed to the real world, due to the significant domain gap between the simulated and real data. In this paper, we propose the first Simulation-to-Reality transfer learning framework for multiagent cooperative perception using a novel Vision Transformer, named as S2R-ViT, which considers both the Implementation Gap and Feature Gap between simulated and real data. We investigate the effects of these two types of domain gaps and propose a novel uncertainty-aware vision transformer to effectively relief the Implementation Gap and an agentbased feature adaptation module with inter-agent and egoagent discriminators to reduce the Feature Gap. Our intensive experiments on the public multi-agent cooperative perception datasets OPV2V and V2V4Real demonstrate that the proposed S2R-ViT can effectively bridge the gap from simulation to reality and outperform other methods significantly for point cloud-based 3D object detection.

show abstract

Section: Quantitaive Evaluationmentioning

confidence: 90%

Section: B Experiments Setupmentioning

confidence: 99%

Decision Support Multi-agent Modeling and Simulation of Aeronautic Marine Oil Spill Response

Liu

Tian

et al. 2023

Smart Innovation, Systems and Technologies

View full text Add to dashboard Cite

show abstract

“…Furthermore, there are domain adaptation methods that aim to address this problem. For example, Li et al 38 designed a domain adaptive object detection framework for autonomous driving under hazy weather to minimize the domain gap between sunny and foggy images at both image and object levels. However, these methods still face challenges in detecting small objects, such as vehicles at long distances on mountain highways due to low-resolution features, resulting in a high missed detection rate.…”

Section: Object Detection In Hazy Motorway Environmentsmentioning

confidence: 99%

“…Furthermore, there are domain adaptation methods that aim to address this problem. For example, Li et al 38 . designed a domain adaptive object detection framework for autonomous driving under hazy weather to minimize the domain gap between sunny and foggy images at both image and object levels.…”

Section: Related Workmentioning

confidence: 99%

Small object detection for autonomous driving under hazy conditions on mountain motorways

Wang,

Qiu,

Jiang

et al. 2023

Opt. Eng.

View full text Add to dashboard Cite

To address the issue of high miss rates for distant small objects and the diminished system detection performance due to the influence of hazy when autonomous vehicles operate on mountain highways. We propose a framework for small object vehicle detection in hazy traffic environments (SHTDet). This framework aims to enhance small object detection for autonomous driving under hazy conditions on mountainous motorways. Specifically, to restore the clarity of hazy images, we designed an image enhancement (IE), and its parameters are predicted by a convolutional neural network [filter parameter estimation (FPE)]. In addition, to enhance the detection accuracy of small objects, we introduce a cascaded sparse query (CSQ) mechanism, which effectively utilizes high-resolution features while maintaining fast detection speed. We jointly optimize the IE and the detection network (CSQ-FCOS) in an end-to-end manner, ensuring that FPE module can learn a suitable IE. Our proposed SHTDet method is adept at adaptively handling sunny and hazy conditions. Extensive experiments demonstrate the efficacy of the SHTDet method in detecting small objects on hazy sections of mountain highways.

show abstract

“…However, fog images with different concentrations have different characteristics, and existing studies tend to ignore this diversity. Previous domain adaptive methods only consider migration learning between the source and target domains while ignoring the joint optimization capability between image defogging, image enhancement, and object detection [ 15 ].…”

Section: Introductionmentioning

confidence: 99%

IDOD-YOLOV7: Image-Dehazing YOLOV7 for Object Detection in Low-Light Foggy Traffic Environments

Qiu

Wang

et al. 2023

Sensors

View full text Add to dashboard Cite

Convolutional neural network (CNN)-based autonomous driving object detection algorithms have excellent detection results on conventional datasets, but the detector performance can be severely degraded in low-light foggy weather environments. Existing methods have difficulty in achieving a balance between low-light image enhancement and object detection. To alleviate this problem, this paper proposes a foggy traffic environment object detection framework, IDOD-YOLOV7. This network is based on joint optimal learning of image defogging module IDOD (AOD + SAIP) and YOLOV7 detection modules. Specifically, for low-light foggy images, we propose to improve the image quality by joint optimization of image defogging (AOD) and image enhancement (SAIP), where the parameters of the SAIP module are predicted by a miniature CNN network and the AOD module performs image defogging by optimizing the atmospheric scattering model. The experimental results show that the IDOD module not only improves the image defogging quality for low-light fog images but also achieves better results in objective evaluation indexes such as PSNR and SSIM. The IDOD and YOLOV7 learn jointly in an end-to-end manner so that object detection can be performed while image enhancement is executed in a weakly supervised manner. Finally, a low-light fogged traffic image dataset (FTOD) was built by physical fogging in order to solve the domain transfer problem. The training of IDOD-YOLOV7 network by a real dataset (FTOD) improves the robustness of the model. We performed various experiments to visually and quantitatively compare our method with several state-of-the-art methods to demonstrate its superiority over the others. The IDOD-YOLOV7 algorithm not only suppresses the artifacts of low-light fog images and improves the visual effect of images but also improves the perception of autonomous driving in low-light foggy environments.

show abstract

Domain Adaptive Object Detection for Autonomous Driving under Foggy Weather

Cited by 64 publications

References 81 publications

Decision Support Multi-agent Modeling and Simulation of Aeronautic Marine Oil Spill Response

Decision Support Multi-agent Modeling and Simulation of Aeronautic Marine Oil Spill Response

Small object detection for autonomous driving under hazy conditions on mountain motorways

IDOD-YOLOV7: Image-Dehazing YOLOV7 for Object Detection in Low-Light Foggy Traffic Environments

Contact Info

Product

Resources

About