FogGuard: guarding YOLO against fog using perceptual loss
- URL: http://arxiv.org/abs/2403.08939v1
- Date: Wed, 13 Mar 2024 20:13:25 GMT
- Title: FogGuard: guarding YOLO against fog using perceptual loss
- Authors: Soheil Gharatappeh, Sepideh Neshatfar, Salimeh Yasaei Sekeh, Vikas Dhiman,
- Abstract summary: FogGuard is a fog-aware object detection network designed to address the challenges posed by foggy weather conditions.
FogGuard achieves 69.43% mAP, as compared to 57.78% for YOLOv3 on the RTTS dataset.
- Score: 5.868532677577194
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: In this paper, we present a novel fog-aware object detection network called FogGuard, designed to address the challenges posed by foggy weather conditions. Autonomous driving systems heavily rely on accurate object detection algorithms, but adverse weather conditions can significantly impact the reliability of deep neural networks (DNNs). Existing approaches fall into two main categories, 1) image enhancement such as IA-YOLO 2) domain adaptation based approaches. Image enhancement based techniques attempt to generate fog-free image. However, retrieving a fogless image from a foggy image is a much harder problem than detecting objects in a foggy image. Domain-adaptation based approaches, on the other hand, do not make use of labelled datasets in the target domain. Both categories of approaches are attempting to solve a harder version of the problem. Our approach builds over fine-tuning on the Our framework is specifically designed to compensate for foggy conditions present in the scene, ensuring robust performance even. We adopt YOLOv3 as the baseline object detection algorithm and introduce a novel Teacher-Student Perceptual loss, to high accuracy object detection in foggy images. Through extensive evaluations on common datasets such as PASCAL VOC and RTTS, we demonstrate the improvement in performance achieved by our network. We demonstrate that FogGuard achieves 69.43\% mAP, as compared to 57.78\% for YOLOv3 on the RTTS dataset. Furthermore, we show that while our training method increases time complexity, it does not introduce any additional overhead during inference compared to the regular YOLO network.
Related papers
- D2SL: Decouple Defogging and Semantic Learning for Foggy Domain-Adaptive Segmentation [0.8261182037130406]
We propose a novel training framework, Decouple Defogging and Semantic learning, called D2SL.
We introduce a domain-consistent transfer strategy to establish a connection between defogging and segmentation tasks.
We design a real fog transfer strategy to improve defogging effects by fully leveraging the fog priors from real foggy images.
arXiv Detail & Related papers (2024-04-07T04:55:58Z) - D-YOLO a robust framework for object detection in adverse weather conditions [0.0]
Adverse weather conditions including haze, snow and rain lead to decline in image qualities, which often causes a decline in performance for deep-learning based detection networks.
To better integrate image restoration and object detection tasks, we designed a double-route network with an attention feature fusion module.
We also proposed a subnetwork to provide haze-free features to the detection network. Specifically, our D-YOLO improves the performance of the detection network by minimizing the distance between the clear feature extraction subnetwork and detection network.
arXiv Detail & Related papers (2024-03-14T09:57:15Z) - DGNet: Dynamic Gradient-Guided Network for Water-Related Optics Image
Enhancement [77.0360085530701]
Underwater image enhancement (UIE) is a challenging task due to the complex degradation caused by underwater environments.
Previous methods often idealize the degradation process, and neglect the impact of medium noise and object motion on the distribution of image features.
Our approach utilizes predicted images to dynamically update pseudo-labels, adding a dynamic gradient to optimize the network's gradient space.
arXiv Detail & Related papers (2023-12-12T06:07:21Z) - DA-RAW: Domain Adaptive Object Detection for Real-World Adverse Weather Conditions [2.048226951354646]
We present an unsupervised domain adaptation framework for object detection in adverse weather conditions.
Our method resolves the style gap by concentrating on style-related information of high-level features.
Using self-supervised contrastive learning, our framework then reduces the weather gap and acquires instance features that are robust to weather corruption.
arXiv Detail & Related papers (2023-09-15T04:37:28Z) - ScatterNeRF: Seeing Through Fog with Physically-Based Inverse Neural
Rendering [83.75284107397003]
We introduce ScatterNeRF, a neural rendering method which renders scenes and decomposes the fog-free background.
We propose a disentangled representation for the scattering volume and the scene objects, and learn the scene reconstruction with physics-inspired losses.
We validate our method by capturing multi-view In-the-Wild data and controlled captures in a large-scale fog chamber.
arXiv Detail & Related papers (2023-05-03T13:24:06Z) - Unpaired Overwater Image Defogging Using Prior Map Guided CycleGAN [60.257791714663725]
We propose a Prior map Guided CycleGAN (PG-CycleGAN) for defogging of images with overwater scenes.
The proposed method outperforms the state-of-the-art supervised, semi-supervised, and unsupervised defogging approaches.
arXiv Detail & Related papers (2022-12-23T03:00:28Z) - GDIP: Gated Differentiable Image Processing for Object-Detection in
Adverse Conditions [15.327704761260131]
We present a Gated Differentiable Image Processing (GDIP) block, a domain-agnostic network architecture.
Our proposed GDIP block learns to enhance images directly through the downstream object detection loss.
We demonstrate significant improvement in detection performance over several state-of-the-art methods.
arXiv Detail & Related papers (2022-09-29T16:43:13Z) - Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions [34.993786158059436]
We propose a novel Image-Adaptive YOLO (IA-YOLO) framework, where each image can be adaptively enhanced for better detection performance.
Specifically, a differentiable image processing (DIP) module is presented to take into account the adverse weather conditions for YOLO detector.
We learn CNN-PP and YOLOv3 jointly in an end-to-end fashion, which ensures CNN-PP can learn an appropriate DIP to enhance the image for detection in a weakly supervised manner.
arXiv Detail & Related papers (2021-12-15T12:54:17Z) - Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in
Adverse Weather [92.84066576636914]
This work addresses the challenging task of LiDAR-based 3D object detection in foggy weather.
We tackle this problem by simulating physically accurate fog into clear-weather scenes.
We are the first to provide strong 3D object detection baselines on the Seeing Through Fog dataset.
arXiv Detail & Related papers (2021-08-11T14:37:54Z) - City-scale Scene Change Detection using Point Clouds [71.73273007900717]
We propose a method for detecting structural changes in a city using images captured from mounted cameras over two different times.
A direct comparison of the two point clouds for change detection is not ideal due to inaccurate geo-location information.
To circumvent this problem, we propose a deep learning-based non-rigid registration on the point clouds.
Experiments show that our method is able to detect scene changes effectively, even in the presence of viewpoint and illumination differences.
arXiv Detail & Related papers (2021-03-26T08:04:13Z) - Camera Invariant Feature Learning for Generalized Face Anti-spoofing [95.30490139294136]
We describe a framework that eliminates the influence of inherent variance from acquisition cameras at the feature level.
Experiments show that the proposed method can achieve better performance in both intra-dataset and cross-dataset settings.
arXiv Detail & Related papers (2021-01-25T13:40:43Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.