Saved in:
Bibliographic Details
Main Authors: Qi, Tianye, Li, Weihao, Barnes, Nick
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.11215
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909957594546176
author Qi, Tianye
Li, Weihao
Barnes, Nick
author_facet Qi, Tianye
Li, Weihao
Barnes, Nick
contents Wildfire smoke is transparent, amorphous, and often visually confounded with clouds, making early-stage detection particularly challenging. In this work, we introduce a benchmark, called SmokeBench, to evaluate the ability of multimodal large language models (MLLMs) to recognize and localize wildfire smoke in images. The benchmark consists of four tasks: (1) smoke classification, (2) tile-based smoke localization, (3) grid-based smoke localization, and (4) smoke detection. We evaluate several MLLMs, including Idefics2, Qwen2.5-VL, InternVL3, Unified-IO 2, Grounding DINO, GPT-4o, and Gemini-2.5 Pro. Our results show that while some models can classify the presence of smoke when it covers a large area, all models struggle with accurate localization, especially in the early stages. Further analysis reveals that smoke volume is strongly correlated with model performance, whereas contrast plays a comparatively minor role. These findings highlight critical limitations of current MLLMs for safety-critical wildfire monitoring and underscore the need for methods that improve early-stage smoke localization.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11215
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SmokeBench: Evaluating Multimodal Large Language Models for Wildfire Smoke Detection
Qi, Tianye
Li, Weihao
Barnes, Nick
Computer Vision and Pattern Recognition
Wildfire smoke is transparent, amorphous, and often visually confounded with clouds, making early-stage detection particularly challenging. In this work, we introduce a benchmark, called SmokeBench, to evaluate the ability of multimodal large language models (MLLMs) to recognize and localize wildfire smoke in images. The benchmark consists of four tasks: (1) smoke classification, (2) tile-based smoke localization, (3) grid-based smoke localization, and (4) smoke detection. We evaluate several MLLMs, including Idefics2, Qwen2.5-VL, InternVL3, Unified-IO 2, Grounding DINO, GPT-4o, and Gemini-2.5 Pro. Our results show that while some models can classify the presence of smoke when it covers a large area, all models struggle with accurate localization, especially in the early stages. Further analysis reveals that smoke volume is strongly correlated with model performance, whereas contrast plays a comparatively minor role. These findings highlight critical limitations of current MLLMs for safety-critical wildfire monitoring and underscore the need for methods that improve early-stage smoke localization.
title SmokeBench: Evaluating Multimodal Large Language Models for Wildfire Smoke Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.11215