# From Filters to VLMs: Benchmarking Defogging Methods through Object Detection and Segmentation Performance

arXiv: https://arxiv.org/abs/2510.03906 · HTML: https://mlsanity.com/paper/2510.03906/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Driving VLMs and open-vocabulary detection, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 35 of its 66 references that are in the map.
- First posted on arXiv 4 Oct 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Driving VLMs and open-vocabulary detection](https://mlsanity.com/mllm/driving-vlms-and-open-vocabulary-detection/)

## Builds on (most cited first, 30 of 35)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space](https://mlsanity.com/paper/2506.15742/) (2025)
- [Panoptic Segmentation](https://mlsanity.com/paper/1801.00868/) (2018)
- [The Perception-Distortion Tradeoff](https://mlsanity.com/paper/1711.06077/) (2017)
- [Semantic Foggy Scene Understanding with Synthetic Data](https://mlsanity.com/paper/1708.07819/) (2017)
- [DehazeNet: An End-to-End System for Single Image Haze Removal](https://mlsanity.com/paper/1601.07661/) (2016)
- [FFA-Net: Feature Fusion Attention Network for Single Image Dehazing](https://mlsanity.com/paper/1911.07559/) (2019)
- [Vision Transformers for Single Image Dehazing](https://mlsanity.com/paper/2204.03883/) (2022)
- [GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing](https://mlsanity.com/paper/1908.03245/) (2019)
- [Prometheus-Vision: Vision-Language Model as a Judge for Fine-Grained Evaluation](https://mlsanity.com/paper/2401.06591/) (2024)
- [Both Style and Fog Matter: Cumulative Domain Adaptation for Semantic Foggy Scene Understanding](https://mlsanity.com/paper/2112.00484/) (2021)
- [End-to-End United Video Dehazing and Detection](https://mlsanity.com/paper/1709.03919/) (2017)
- [ACDC: The Adverse Conditions Dataset with Correspondences for Robust Semantic Driving Scene Perception](https://mlsanity.com/paper/2104.13395/) (2021)
- [Top-H Decoding: Adapting the Creativity and Coherence with Bounded Entropy in Text Generation](https://mlsanity.com/paper/2509.02510/) (2025)
- [FogAdapt: Self-Supervised Domain Adaptation for Semantic Segmentation of Foggy Images](https://mlsanity.com/paper/2201.02588/) (2022)
- [Convolutional Learning on Directed Acyclic Graphs](https://mlsanity.com/paper/2405.03056/) (2024)
- [VISTA: Vision-Language Inference for Training-Free Stock Time-Series Analysis](https://mlsanity.com/paper/2505.18570/) (2025)
- [Advanced Predictive Modeling for Enhanced Mortality Prediction in ICU Stroke Patients Using Clinical Data](https://mlsanity.com/paper/2407.14211/) (2024)
- [Optimizing Mortality Prediction for ICU Heart Failure Patients: Leveraging XGBoost and Advanced Machine Learning with the MIMIC-III Database](https://mlsanity.com/paper/2409.01685/) (2024)
- [Large Language Model-based Nonnegative Matrix Factorization For Cardiorespiratory Sound Separation](https://mlsanity.com/paper/2502.05757/) (2025)

## Built on by (most cited first, 3 of 3)

- [NeuroMambaLLM: Dynamic Graph Learning of fMRI Functional Connectivity in Autistic Brains Using Mamba and Language Model Reasoning](https://mlsanity.com/paper/2602.13770/) (2026)
- [Can MLLMs Critique Like Humans? Evaluating Open-Ended Aesthetic Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2606.29689/) (2026)
- [CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits](https://mlsanity.com/paper/2608.05732/) (2026)

## Abstract (arXiv)

Autonomous driving perception systems are particularly vulnerable in foggy conditions, where light scattering reduces contrast and obscures fine details critical for safe operation. While numerous defogging methods exist, from handcrafted filters to learned restoration models, improvements in image fidelity do not consistently translate into better downstream detection and segmentation. Moreover, prior evaluations often rely on synthetic data, raising concerns about real-world transferability. We present a structured empirical study that benchmarks a comprehensive set of defogging pipelines, including classical dehazing filters, modern defogging networks, chained variants combining filters and models, and prompt-driven visual language image editing models applied directly to foggy images. To bridge the gap between simulated and physical environments, we evaluate these pipelines on both the synthetic Foggy Cityscapes dataset and the real-world Adverse Conditions Dataset with Correspondences (ACDC). We examine generalization by evaluating performance on synthetic fog and real-world conditions, assessing both image quality and downstream perception in terms of object detection mean average precision and segmentation panoptic quality. Our analysis identifies when defogging is effective, the impact of combining models, and how visual language models compare to traditional approaches. We additionally report qualitative rubric-based evaluations from both human and visual language model judges and analyze their alignment with downstream task metrics. Together, these results establish a transparent, task-oriented benchmark for defogging methods and identify the conditions under which pre-processing meaningfully improves autonomous perception in adverse weather. Project page: https://aradfir.github.io/filters-to-vlms-defogging-page/