English
Related papers

Related papers: ToSA: Token Merging with Spatial Awareness

200 papers

In this study, we introduce a novel method called group-wise \textbf{VI}sual token \textbf{S}election and \textbf{A}ggregation (VISA) to address the issue of inefficient inference stemming from excessive visual tokens in multimoal large…

Computer Vision and Pattern Recognition · Computer Science 2025-08-26 Pengfei Jiang , Hanjun Li , Linglan Zhao , Fei Chao , Ke Yan , Shouhong Ding , Rongrong Ji

Slow inference speed is one of the most crucial concerns for deploying multi-view 3D detectors to tasks with high real-time requirements like autonomous driving. Although many sparse query-based methods have already attempted to improve the…

Computer Vision and Pattern Recognition · Computer Science 2024-09-04 Dingyuan Zhang , Dingkang Liang , Zichang Tan , Xiaoqing Ye , Cheng Zhang , Jingdong Wang , Xiang Bai

As the scale of data and models for video understanding rapidly expand, handling long-form video input in transformer-based models presents a practical challenge. Rather than resorting to input sampling or token dropping, which may result…

Computer Vision and Pattern Recognition · Computer Science 2024-11-01 Seon-Ho Lee , Jue Wang , Zhikang Zhang , David Fan , Xinyu Li

Large Multimodal Models (LMMs) have shown significant visual reasoning capabilities by connecting a visual encoder and a large language model. LMMs typically take in a fixed and large amount of visual tokens, such as the penultimate layer…

Computer Vision and Pattern Recognition · Computer Science 2026-02-03 Yuzhang Shang , Mu Cai , Bingxin Xu , Yong Jae Lee , Yan Yan

Increasing the throughput of the Transformer architecture, a foundational component used in numerous state-of-the-art models for vision and language tasks (e.g., GPT, LLaVa), is an important problem in machine learning. One recent and…

Weakly supervised object localization (WSOL) is a challenging task aiming to localize objects with only image-level supervision. Recent works apply visual transformer to WSOL and achieve significant success by exploiting the long-range…

Computer Vision and Pattern Recognition · Computer Science 2023-08-10 Pingyu Wu , Wei Zhai , Yang Cao , Jiebo Luo , Zheng-Jun Zha

Multi-head-self-attention (MHSA)-equipped models have achieved notable performance in computer vision. Their computational complexity is proportional to quadratic numbers of pixels in input feature maps, resulting in slow processing,…

Computer Vision and Pattern Recognition · Computer Science 2023-12-19 Yuki Tatsunami , Masato Taki

Recent token reduction methods for Vision Transformers (ViTs) incorporate token merging, which measures the similarities between token embeddings and combines the most similar pairs. However, their merging policies are directly dependent on…

Computer Vision and Pattern Recognition · Computer Science 2024-12-17 Dong Hoon Lee , Seunghoon Hong

Vision Transformers can achieve high accuracy and strong generalization across various contexts, but their practical applicability on real-world robotic systems is limited due to their quadratic attention complexity. Recent works have…

Computer Vision and Pattern Recognition · Computer Science 2026-05-04 Fabio Montello , Ronja Güldenring , Lazaros Nalpantidis

Recent Vision Transformer~(ViT) models have demonstrated encouraging results across various computer vision tasks, thanks to their competence in modeling long-range dependencies of image patches or tokens via self-attention. These models,…

Computer Vision and Pattern Recognition · Computer Science 2022-04-14 Sucheng Ren , Daquan Zhou , Shengfeng He , Jiashi Feng , Xinchao Wang

Feature shifts have been shown to be useful for action recognition with CNN-based models since Temporal Shift Module (TSM) was proposed. It is based on frame-wise feature extraction with late fusion, and layer features are shifted along the…

Computer Vision and Pattern Recognition · Computer Science 2022-11-15 Ryota Hashiguchi , Toru Tamaki

Utilizing transformer architectures for semantic segmentation of high-resolution images is hindered by the attention's quadratic computational complexity in the number of tokens. A solution to this challenge involves decreasing the number…

Computer Vision and Pattern Recognition · Computer Science 2024-05-24 Daniel Kienzle , Marco Kantonis , Robin Schön , Rainer Lienhart

Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant…

Robotics · Computer Science 2025-12-11 Yifan Ye , Jiaqi Ma , Jun Cen , Zhihe Lu

Although text-to-image (T2I) models exhibit remarkable generation capabilities, they frequently fail to accurately bind semantically related objects or attributes in the input prompts; a challenge termed semantic binding. Previous…

Computer Vision and Pattern Recognition · Computer Science 2024-11-12 Taihang Hu , Linxuan Li , Joost van de Weijer , Hongcheng Gao , Fahad Shahbaz Khan , Jian Yang , Ming-Ming Cheng , Kai Wang , Yaxing Wang

Semi-supervised semantic segmentation has witnessed remarkable advancements in recent years. However, existing algorithms are based on convolutional neural networks and directly applying them to Vision Transformers poses certain limitations…

Computer Vision and Pattern Recognition · Computer Science 2025-03-11 Dengke Zhang , Quan Tang , Fagui Liu , Haiqing Mei , C. L. Philip Chen

Vision Transformer (ViT) architectures represent images as collections of high-dimensional vectorized tokens, each corresponding to a rectangular non-overlapping patch. This representation trades spatial granularity for embedding…

Computer Vision and Pattern Recognition · Computer Science 2024-05-08 Dong Lao , Yangchao Wu , Tian Yu Liu , Alex Wong , Stefano Soatto

In this paper, we observe two levels of redundancies when applying vision transformers (ViT) for image recognition. First, fixing the number of tokens through the whole network produces redundant features at the spatial level. Second, the…

Computer Vision and Pattern Recognition · Computer Science 2021-08-10 Boyu Chen , Peixia Li , Baopu Li , Chuming Li , Lei Bai , Chen Lin , Ming Sun , Junjie Yan , Wanli Ouyang

Due to spatial redundancy in remote sensing images, sparse tokens containing rich information are usually involved in self-attention (SA) to reduce the overall token numbers within the calculation, avoiding the high computational cost issue…

Computer Vision and Pattern Recognition · Computer Science 2024-05-17 Wentao Jiang , Jing Zhang , Di Wang , Qiming Zhang , Zengmao Wang , Bo Du

In this paper, we present a new approach for model acceleration by exploiting spatial sparsity in visual data. We observe that the final prediction in vision Transformers is only based on a subset of the most informative tokens, which is…

Computer Vision and Pattern Recognition · Computer Science 2023-06-05 Yongming Rao , Zuyan Liu , Wenliang Zhao , Jie Zhou , Jiwen Lu

The quadratic cost of self-attention in Vision Transformers (ViTs) constitutes a fundamental bottleneck for practical deployment, motivating a vibrant line of research on token reduction. Among existing approaches, token merging (ToMe) has…

Computer Vision and Pattern Recognition · Computer Science 2026-05-28 Semi Lee , Hyejin Go , Hyesong Choi