Globality Strikes Back: Rethinking the Global Knowledge of CLIP in Training-Free Open-Vocabulary Semantic Segmentation
- URL: http://arxiv.org/abs/2502.06818v1
- Date: Wed, 05 Feb 2025 03:37:50 GMT
- Title: Globality Strikes Back: Rethinking the Global Knowledge of CLIP in Training-Free Open-Vocabulary Semantic Segmentation
- Authors: Jingyun Wang, Cilin Yan, Guoliang Kang,
- Abstract summary: Recent works modify CLIP to perform semantic segmentation in a training-free manner (TF-OVSS)
With their modifications, the ability of CLIP to aggregate global context information is largely weakened.
We propose a new method named GCLIP, which mines the beneficial global knowledge of CLIP to facilitate the TF-OVSS task.
- Score: 5.3499687969383345
- License:
- Abstract: Recent works modify CLIP to perform open-vocabulary semantic segmentation in a training-free manner (TF-OVSS). In CLIP, patch-wise image representations mainly encode the homogeneous image-level properties and thus are not discriminative enough, hindering its application to the dense prediction task. Previous works make image features more distinct across patches, through making each patch mainly attend to itself or the neighboring patches within a narrow local window. However, with their modifications, the ability of CLIP to aggregate global context information, which is known to be useful for distinguishing confusing categories, is largely weakened. In this paper, we propose a new method named GCLIP, which mines the beneficial global knowledge of CLIP to facilitate the TF-OVSS task. Firstly, we aim to equip the last-block attention with image-level properties while not introducing homogeneous attention patterns across patches. In GCLIP, we merge the attention from the global token emerging blocks with the Query-Query attention to realize this goal. Secondly, we aim to make the Value embeddings of the last-block attention module more distinct and semantically correlated. To realize this, we design a novel channel suppression strategy. As the representation of each patch is finally determined by the attention weights and the Value embeddings, our method can generate more discriminative patch-level image features while absorbing global context information. Extensive experiments on five standard benchmarks demonstrate that our method consistently outperforms previous state-of-the-arts.
Related papers
- ResCLIP: Residual Attention for Training-free Dense Vision-language Inference [27.551367463011008]
Cross-correlation of self-attention in CLIP's non-final layers also exhibits localization properties.
We propose the Residual Cross-correlation Self-attention (RCS) module, which leverages the cross-correlation self-attention from intermediate layers to remold the attention in the final block.
The RCS module effectively reorganizes spatial information, unleashing the localization potential within CLIP for dense vision-language inference.
arXiv Detail & Related papers (2024-11-24T14:14:14Z) - Point Cloud Understanding via Attention-Driven Contrastive Learning [64.65145700121442]
Transformer-based models have advanced point cloud understanding by leveraging self-attention mechanisms.
PointACL is an attention-driven contrastive learning framework designed to address these limitations.
Our method employs an attention-driven dynamic masking strategy that guides the model to focus on under-attended regions.
arXiv Detail & Related papers (2024-11-22T05:41:00Z) - Towards Open-Vocabulary Semantic Segmentation Without Semantic Labels [53.8817160001038]
We propose a novel method, PixelCLIP, to adapt the CLIP image encoder for pixel-level understanding.
To address the challenges of leveraging masks without semantic labels, we devise an online clustering algorithm.
PixelCLIP shows significant performance improvements over CLIP and competitive results compared to caption-supervised methods.
arXiv Detail & Related papers (2024-09-30T01:13:03Z) - DIAL: Dense Image-text ALignment for Weakly Supervised Semantic Segmentation [8.422110274212503]
Weakly supervised semantic segmentation approaches typically rely on class activation maps (CAMs) for initial seed generation.
We introduce DALNet, which leverages text embeddings to enhance the comprehensive understanding and precise localization of objects across different levels of granularity.
Our approach, in particular, allows for more efficient end-to-end process as a single-stage method.
arXiv Detail & Related papers (2024-09-24T06:51:49Z) - Explore the Potential of CLIP for Training-Free Open Vocabulary Semantic Segmentation [38.16802763051431]
We propose CLIPtrase, a training-free semantic segmentation strategy.
It enhances local feature awareness through recalibrated self-correlation among patches.
Experiments show that we are 22.3% ahead of CLIP on average on 9 segmentation benchmarks.
arXiv Detail & Related papers (2024-07-11T08:12:16Z) - LDCA: Local Descriptors with Contextual Augmentation for Few-Shot
Learning [0.0]
We introduce a novel approach termed "Local Descriptor with Contextual Augmentation (LDCA)"
LDCA bridges the gap between local and global understanding by leveraging an adaptive global contextual enhancement module.
Experiments underscore the efficacy of our method, showing a maximal absolute improvement of 20% over the next-best on fine-grained classification datasets.
arXiv Detail & Related papers (2024-01-24T14:44:48Z) - UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding [90.74967596080982]
This paper extends Contrastive Language-Image Pre-training (CLIP) with multi-granularity alignment.
We develop a Unified Multi-Granularity learning framework, termed UMG-CLIP, which simultaneously empowers the model with versatile perception abilities.
With parameter efficient tuning, UMG-CLIP surpasses current widely used CLIP variants and achieves state-of-the-art performance on diverse image understanding benchmarks.
arXiv Detail & Related papers (2024-01-12T06:35:09Z) - TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary
Multi-Label Classification of CLIP Without Training [29.431698321195814]
Contrastive Language-Image Pre-training (CLIP) has demonstrated impressive capabilities in open-vocabulary classification.
CLIP shows poor performance on multi-label datasets because the global feature tends to be dominated by the most prominent class.
We propose a local-to-global framework to obtain image tags.
arXiv Detail & Related papers (2023-12-20T08:15:40Z) - TagCLIP: Improving Discrimination Ability of Open-Vocabulary Semantic Segmentation [53.974228542090046]
Contrastive Language-Image Pre-training (CLIP) has recently shown great promise in pixel-level zero-shot learning tasks.
Existing approaches utilizing CLIP's text and patch embeddings to generate semantic masks often misidentify input pixels from unseen classes.
We propose TagCLIP (Trusty-aware guided CLIP) to address this issue.
arXiv Detail & Related papers (2023-04-15T12:52:23Z) - Global Knowledge Calibration for Fast Open-Vocabulary Segmentation [124.74256749281625]
We introduce a text diversification strategy that generates a set of synonyms for each training category.
We also employ a text-guided knowledge distillation method to preserve the generalizable knowledge of CLIP.
Our proposed model achieves robust generalization performance across various datasets.
arXiv Detail & Related papers (2023-03-16T09:51:41Z) - No Token Left Behind: Explainability-Aided Image Classification and
Generation [79.4957965474334]
We present a novel explainability-based approach, which adds a loss term to ensure that CLIP focuses on all relevant semantic parts of the input.
Our method yields an improvement in the recognition rate, without additional training or fine-tuning.
arXiv Detail & Related papers (2022-04-11T07:16:39Z)
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.