ClearVision: Leveraging CycleGAN and SigLIP-2 for Robust All-Weather Classification in Traffic Camera Imagery

• URL: http://arxiv.org/abs/2504.19684v1
• Date: Mon, 28 Apr 2025 11:22:08 GMT
• Title: ClearVision: Leveraging CycleGAN and SigLIP-2 for Robust All-Weather Classification in Traffic Camera Imagery
• Abstract summary: We propose a scalable framework that combines generative domain adaptation with efficient contrastive learning to enhance classification performance.<n>Using CycleGAN-based domain translation, we improve the quality of nighttime images, enabling better feature extraction by downstream models.
• Score: 1.9953114232105387
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate weather classification from low-quality traffic camera imagery remains a challenging task, particularly under adverse nighttime conditions. In this study, we propose a scalable framework that combines generative domain adaptation with efficient contrastive learning to enhance classification performance. Using CycleGAN-based domain translation, we improve the quality of nighttime images, enabling better feature extraction by downstream models. While the baseline EVA-02 model employing CLIP-based contrastive loss achieves an overall accuracy of 96.55\%, it exhibits a significant performance gap between daytime (97.21\%) and nighttime conditions (63.40\%). Replacing CLIP with the lightweight SigLIP-2 (Sigmoid contrastive loss) achieves a competitive overall accuracy of 94.00\%, with substantial improvements in nighttime performance (85.90\% accuracy). The combination of Vision-SigLIP-2, Text-SigLIP-2, CycleGAN, and contrastive training achieves the best nighttime accuracy (85.90\%) among all models tested, while EVA-02 with CycleGAN maintains the highest overall accuracy (97.01\%) and per-class accuracies. These findings demonstrate the potential of combining domain adaptation and efficient contrastive learning to build practical, resource-efficient weather classification systems for intelligent transportation infrastructure.

Related papers

• Calibration-Free Vehicle Speed Estimation: A Monocular Keypoint-Template Approach [0.8819673391477033]
  This paper proposes a calibration-free framework for reliably and effectively estimating vehicle speeds from monocular videos.<n>The proposed framework estimates vehicle speeds using a 36-keypoint vehicle template and a homography matrix updated at each frame.
  arXiv  Detail & Related papers  (2026-08-17T16:39:06Z)
• StableTTA: Training-Free Test-Time Adaptation that Improves Model Accuracy on ImageNet1K to 96% [4.467218412104476]
  Ensemble methods are widely used to improve predictive performance, but their effectiveness often comes at the cost of increased memory usage and computational complexity.<n>We propose test-time adaptation (StableTTA), a training-free method employs novel image and logit processing.
  arXiv  Detail & Related papers  (2026-04-06T09:21:48Z)
• HG-Lane: High-Fidelity Generation of Lane Scenes under Adverse Weather and Lighting Conditions without Re-annotation [11.667208452607959]
  We propose HG-Lane, a High-fidelity Generation framework for Lane Scenes under adverse weather and lighting conditions.<n> Experimental results demonstrate that our method consistently and significantly improves the performance of existing lane detection networks.
  arXiv  Detail & Related papers  (2026-03-10T18:04:35Z)
• NDLPNet: A Location-Aware Nighttime Deraining Network and a Real-World Benchmark Dataset [8.582528726118023]
  Rain streak artifacts hamper the performance of nighttime surveillance and autonomous navigation.<n>We propose a novel Nighttime Deraining Location-enhanced Perceptual Network (NDLPNet)<n>NDLPNet captures the spatial positional information and density distribution of rain streaks in low-light environments.
  arXiv  Detail & Related papers  (2025-09-17T07:24:47Z)
• PRE-Mamba: A 4D State Space Model for Ultra-High-Frequent Event Camera Deraining [47.81253972389206]
  Event cameras excel in high temporal resolution and dynamic range but suffer from dense noise in rainy conditions.<n>We propose PRE-Mamba, a novel point-based camera framework for event deraining.
  arXiv  Detail & Related papers  (2025-05-08T14:52:45Z)
• Segregation and Context Aggregation Network for Real-time Cloud Segmentation [16.33805395655703]
  Ground-based sky/cloud segmentation extracts clouds for further feature analysis.<n>We introduce SCANet, a novel lightweight cloud segmentation model.<n> SCANet achieves state-of-the-art performance while drastically reducing computational complexity.
  arXiv  Detail & Related papers  (2025-04-19T04:47:59Z)
• Speedy MASt3R [68.47052557089631]
  MASt3R redefines image matching as a 3D task by leveraging DUSt3R and introducing a fast reciprocal matching scheme.<n>Fast MASt3R achieves a 54% reduction in inference time (198 ms to 91 ms per image pair) without sacrificing accuracy.<n>This advancement enables real-time 3D understanding, benefiting applications like mixed reality navigation and large-scale 3D scene reconstruction.
  arXiv  Detail & Related papers  (2025-03-13T03:56:22Z)
• CLIPure: Purification in Latent Space via CLIP for Adversarially Robust Zero-Shot Classification [65.46685389276443]
  We ground our work on CLIP, a vision-language pre-trained encoder model that can perform zero-shot classification by matching an image with text prompts. We then formulate purification risk as the KL divergence between the joint distributions purification process. We propose two variants for our CLIPure approach: CLI-Diff which models the likelihood of images' latent vectors, and CLIPure-Cos which models the likelihood with the cosine similarity between the embeddings of an image and a photo of a.''
  arXiv  Detail & Related papers  (2025-02-25T13:09:34Z)
• WARLearn: Weather-Adaptive Representation Learning [4.5035146256368455]
  WARLearn is a novel framework designed for adaptive representation learning in challenging and adversarial weather conditions. Our method exhibits remarkable performance gains in scenarios characterized by fog and low-light conditions.
  arXiv  Detail & Related papers  (2024-11-21T13:00:30Z)
• Real-Time Weather Image Classification with SVM [3.842565887299043]
  This paper presents a study on classifying weather conditions in images into four categories: rainy, low light, haze, and clear. The motivation for this work stems from the need to improve the reliability and efficiency of automated systems.
  arXiv  Detail & Related papers  (2024-09-01T19:41:35Z)
• An Experimental Study on Exploring Strong Lightweight Vision Transformers via Masked Image Modeling Pre-Training [51.622652121580394]
  Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the textitextremely simple lightweight ViTs' fine-tuning performance can also benefit from this pre-training paradigm. Our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design ($5.7M$/$6.5M$) can achieve $79.4%$/$78.9%$ top-1 accuracy on ImageNet-1
  arXiv  Detail & Related papers  (2024-04-18T14:14:44Z)
• Hierarchical-level rain image generative model based on GAN [4.956959291938016]
  hierarchical-level rain image generative model, rain conditional CycleGAN, is constructed. Different rain intensities are introduced as labels in conditional GAN. Model structure is optimized and the training strategy is adjusted to alleviate the problem of mode collapse.
  arXiv  Detail & Related papers  (2023-09-06T12:59:52Z)
• Improving Panoptic Segmentation for Nighttime or Low-Illumination Urban Driving Scenes [0.0]
  We propose two new methods to improve the performance and robustness of Panoptic segmentation. One of the main factors for poor results is the lack of sufficient and accurately annotated nighttime images for urban driving scenes. The proposed approach makes use of CycleGAN to translate daytime images with existing panoptic annotations into nighttime images. It is then utilized to retrain a Panoptic segmentation model to improve performance and robustness under poor illumination and nighttime conditions.
  arXiv  Detail & Related papers  (2023-06-23T18:14:26Z)
• Patch-Level Contrasting without Patch Correspondence for Accurate and Dense Contrastive Representation Learning [79.43940012723539]
  ADCLR is a self-supervised learning framework for learning accurate and dense vision representation. Our approach achieves new state-of-the-art performance for contrastive methods.
  arXiv  Detail & Related papers  (2023-06-23T07:38:09Z)
• LUT-GCE: Lookup Table Global Curve Estimation for Fast Low-light Image Enhancement [62.17015413594777]
  We present an effective and efficient approach for low-light image enhancement, named LUT-GCE. We estimate a global curve for the entire image that allows corrections for both under- and over-exposure. Our approach outperforms the state of the art in terms of inference speed, especially on high-definition images (e.g., 1080p and 4k)
  arXiv  Detail & Related papers  (2023-06-12T12:53:06Z)
• Advancing Unsupervised Low-light Image Enhancement: Noise Estimation, Illumination Interpolation, and Self-Regulation [55.07472635587852]
  Low-Light Image Enhancement (LLIE) techniques have made notable advancements in preserving image details and enhancing contrast. These approaches encounter persistent challenges in efficiently mitigating dynamic noise and accommodating diverse low-light scenarios. We first propose a method for estimating the noise level in low light images in a quick and accurate way. We then devise a Learnable Illumination Interpolator (LII) to satisfy general constraints between illumination and input.
  arXiv  Detail & Related papers  (2023-05-17T13:56:48Z)
• Getting More Juice Out of Your Data: Hard Pair Refinement Enhances Visual-Language Models Without Extra Data [122.282521548393]
  Contrastive Language-Image Pre-training (CLIP) has become the standard for cross-modal image-text representation learning.<n>We introduce HELIP, a cost-effective strategy that improves CLIP models by exploiting challenging text-image pairs within existing datasets in continuous training.
  arXiv  Detail & Related papers  (2023-05-09T07:00:17Z)
• CLIP Itself is a Strong Fine-tuner: Achieving 85.7% and 88.0% Top-1 Accuracy with ViT-B and ViT-L on ImageNet [139.56863124214905]
  We find that fine-tuning performance of CLIP is substantially underestimated. Specifically, CLIP ViT-Base/16 and CLIP ViT-Large/14 can achieve 85.7%,88.0% finetuning Top-1 accuracy on the ImageNet-1K dataset.
  arXiv  Detail & Related papers  (2022-12-12T18:59:59Z)
• When the Sun Goes Down: Repairing Photometric Losses for All-Day Depth Estimation [47.617222712429026]
  We show how to use a combination of three techniques to allow the existing photometric losses to work for both day and nighttime images. First, we introduce a per-pixel neural intensity transformation to compensate for the light changes that occur between successive frames. Second, we predict a per-pixel residual flow map that we use to correct the reprojection correspondences induced by the estimated ego-motion and depth.
  arXiv  Detail & Related papers  (2022-06-28T09:29:55Z)
• Toward Fast, Flexible, and Robust Low-Light Image Enhancement [87.27326390675155]
  We develop a new Self-Calibrated Illumination (SCI) learning framework for fast, flexible, and robust brightening images in real-world low-light scenarios. Considering the computational burden of the cascaded pattern, we construct the self-calibrated module which realizes the convergence between results of each stage. We make comprehensive explorations to SCI's inherent properties including operation-insensitive adaptability and model-irrelevant generality.
  arXiv  Detail & Related papers  (2022-04-21T14:40:32Z)
• Provable Stochastic Optimization for Global Contrastive Learning: Small Batch Does Not Harm Performance [53.49803579981569]
  We consider a global objective for contrastive learning, which contrasts each positive pair with all negative pairs for an anchor point. Existing methods such as SimCLR requires a large batch size in order to achieve a satisfactory result. We propose a memory-efficient optimization algorithm for solving the Global Contrastive Learning of Representations, named SogCLR.
  arXiv  Detail & Related papers  (2022-02-24T22:16:53Z)
• TransWeather: Transformer-based Restoration of Images Degraded by Adverse Weather Conditions [77.20136060506906]
  We propose TransWeather, a transformer-based end-to-end model with just a single encoder and a decoder. TransWeather achieves significant improvements across multiple test datasets over both All-in-One network. It is validated on real world test images and found to be more effective than previous methods.
  arXiv  Detail & Related papers  (2021-11-29T18:57:09Z)
• MIO : Mutual Information Optimization using Self-Supervised Binary Contrastive Learning [12.365801596593936]
  We model our pre-training task as a binary classification problem to induce an implicit contrastive effect.<n>Unlike existing methods, the proposed loss function optimize the mutual information in positive and negative pairs.<n>The proposed method outperforms SOTA self-supervised contrastive frameworks on benchmark datasets.
  arXiv  Detail & Related papers  (2021-11-24T17:51:29Z)
• Bi-Mix: Bidirectional Mixing for Domain Adaptive Nighttime Semantic Segmentation [83.97914777313136]
  In autonomous driving, learning a segmentation model that can adapt to various environmental conditions is crucial. In this paper, we study the problem of Domain Adaptive Nighttime Semantic (DANSS), which aims to learn a discriminative nighttime model. We propose a novel Bi-Mix framework for DANSS, which can contribute to both image translation and segmentation adaptation processes.
  arXiv  Detail & Related papers  (2021-11-19T17:39:47Z)

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.