Salvato in:
Dettagli Bibliografici
Autori principali: Xue, Leyan, Han, Zongbo, Wang, Guangyu, Hu, Qinghua, Cheng, Mingyue, Zhang, Changqing
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2507.03458
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915371890507776
author Xue, Leyan
Han, Zongbo
Wang, Guangyu
Hu, Qinghua
Cheng, Mingyue
Zhang, Changqing
author_facet Xue, Leyan
Han, Zongbo
Wang, Guangyu
Hu, Qinghua
Cheng, Mingyue
Zhang, Changqing
contents Vision-Language Models (VLMs) like CLIP achieve cross-modal semantic alignment through contrastive learning, exhibiting robust zero-shot generalization. Traditional prompt engineering, however, predominantly relies on coarse-grained category labels, neglecting fine-grained local semantics. Existing approaches assume that VLMs inherently recognize localized visual details and attempt to enhance classification by augmenting text prompts with attribute descriptors generated by large language models. However, our systematic experiments reveal critical limitations: CLIP's strong bias toward global image patterns hinders its ability to process localized visual descriptors. To address this fundamental constraint, we propose a simple, effective, and plug-and-play solution that enables CLIP to ``See Both the Forest and the Trees." Specifically, we employ stochastic multi-crop augmentation to activate CLIP's latent capacity for localized feature analysis. By cropping only partial regions, the approach effectively constrains the model's receptive field and recalibrates its attention mechanism, thereby mitigating its inherent bias. We evaluate the proposed method under zero-shot, few-shot, and test-time adaptation settings, and extensive experiments demonstrate that D&D achieves promising performance.
format Preprint
id arxiv_https___arxiv_org_abs_2507_03458
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Helping CLIP See Both the Forest and the Trees: A Decomposition and Description Approach
Xue, Leyan
Han, Zongbo
Wang, Guangyu
Hu, Qinghua
Cheng, Mingyue
Zhang, Changqing
Computer Vision and Pattern Recognition
Artificial Intelligence
Vision-Language Models (VLMs) like CLIP achieve cross-modal semantic alignment through contrastive learning, exhibiting robust zero-shot generalization. Traditional prompt engineering, however, predominantly relies on coarse-grained category labels, neglecting fine-grained local semantics. Existing approaches assume that VLMs inherently recognize localized visual details and attempt to enhance classification by augmenting text prompts with attribute descriptors generated by large language models. However, our systematic experiments reveal critical limitations: CLIP's strong bias toward global image patterns hinders its ability to process localized visual descriptors. To address this fundamental constraint, we propose a simple, effective, and plug-and-play solution that enables CLIP to ``See Both the Forest and the Trees." Specifically, we employ stochastic multi-crop augmentation to activate CLIP's latent capacity for localized feature analysis. By cropping only partial regions, the approach effectively constrains the model's receptive field and recalibrates its attention mechanism, thereby mitigating its inherent bias. We evaluate the proposed method under zero-shot, few-shot, and test-time adaptation settings, and extensive experiments demonstrate that D&D achieves promising performance.
title Helping CLIP See Both the Forest and the Trees: A Decomposition and Description Approach
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2507.03458