检索结果-内蒙古大学图书馆

International Conference on computer vision Workshops (ICCV Workshops)

作者： Ce Zheng Matias Mendieta Chen Chen Center for Research in Computer Vision University of Central Florida

Facial expression recognition (FER) is an important task in computer vision, having practical applications in areas such as human-computer interaction, education, health-care, and online monitoring. In this challenging FER task, there are three key issues especially prevalent: inter-class similarity, intra-class discrepancy, and scale sensitivity. While existing works typically address some of these issues, none have fully addressed all three challenges in a unified framework. In this paper, we propose a two-stream Pyramid crOss-fuSion TransformER network (POSTER), that aims to holistically solve all three issues. Specifically, we design a transformer-based cross-fusion method that enables effective collaboration of facial landmark features and image features to maximize proper attention to salient facial regions. Furthermore, POSTER employs a pyramid structure to promote scale invariance. Extensive experimental results demonstrate that our POSTER achieves new state-of-the-art results on RAF-DB (92.05%), FERPlus (91.62%), as well as AffectNet 7 class (67.31%) and 8 class (63.34%). Code is available at https://***/zczcwh/POSTER.

关键词：

来源：评论

学校读者我要写书评

暂无评论

HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video Understanding

arXiv

引用

arXiv 2025年

作者： Azad, Shehreen Vineet, Vibhav Rawat, Yogesh Singh Center for Research in Computer Vision University of Central Florida United States Microsoft Research United States

Despite advancements in multimodal large language models (MLLMs), current approaches struggle in medium-to-long video understanding due to frame and context length limitations. As a result, these models often depend on frame sampling, which risks missing key information over time and lacks task-specific relevance. To address these challenges, we introduce HierarQ, a task-aware hierarchical Q-Former based framework that sequentially processes frames to bypass the need for frame sampling, while avoiding LLM’s context length limitations. We introduce a lightweight two-stream language-guided feature modulator to incorporate task awareness in video understanding, with the entity stream capturing frame-level object information within a short context and the scene stream identifying their broader interactions over longer period of time. Each stream is supported by dedicated memory banks which enables our proposed Hierarchical Querying transformer (HierarQ) to effectively capture short and long-term context. Extensive evaluations on 10 video benchmarks across video understanding, question answering, and captioning tasks demonstrate HierarQ’s state-of-the-art performance across most datasets, proving its robustness and efficiency for comprehensive video analysis. © 2025, CC BY.

关键词： Structured Query Language

来源：评论

学校读者我要写书评

暂无评论

GeoCLIP: clip-inspired alignment between locations and images for effective worldwide geo-localization 23

GeoCLIP: clip-inspired alignment between locations and image...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Vicente Vivanco Cepeda Gaurav Kumar Nayak Mubarak Shah Center for Research in Computer Vision University of Central Florida

Worldwide Geo-localization aims to pinpoint the precise location of images taken anywhere on Earth. This task has considerable challenges due to the immense variation in geographic landscapes. The image-to-image retrieval-based approaches fail to solve this problem on a global scale as it is not feasible to construct a large gallery of images covering the entire world. Instead, existing approaches divide the globe into discrete geographic cells, transforming the problem into a classification task. However, their performance is limited by the predefined classes and often results in inaccurate localizations when an image's location significantly deviates from its class center. To overcome these limitations, we propose GeoCLIP, a novel CLIP-inspired Image-to-GPS retrieval approach that enforces alignment between the image and its corresponding GPS locations. GeoCLIP's location encoder models the Earth as a continuous function by employing positional encoding through random Fourier features and constructing a hierarchical representation that captures information at varying resolutions to yield a semantically rich high-dimensional feature suitable to use even beyond geo-localization. To the best of our knowledge, this is the first work employing GPS encoding for geo-localization. We demonstrate the efficacy of our method via extensive experiments and ablations on benchmark datasets. We achieve competitive performance with just 20% of training data, highlighting its effectiveness even in limited-data settings. Furthermore, we qualitatively demonstrate geo-localization using a text query by leveraging the CLIP backbone of our image encoder. The project webpage is available at: https://***/GeoCLIP

关键词：

来源：评论

学校读者我要写书评

暂无评论

Source-free Domain Adaptive Human Pose Estimation

Source-free Domain Adaptive Human Pose Estimation

引用

International Conference on computer vision (ICCV)

作者： Qucheng Peng Ce Zheng Chen Chen Center for Research in Computer Vision University of Central Florida

Human Pose Estimation (HPE) is widely used in various fields, including motion analysis, healthcare, and virtual reality. However, the great expenses of labeled real-world datasets present a significant challenge for HPE. To overcome this, one approach is to train HPE models on synthetic datasets and then perform domain adaptation (DA) on real-world data. Unfortunately, existing DA methods for HPE neglect data privacy and security by using both source and target data in the adaptation *** this end, we propose a new task, named source-free domain adaptive HPE, which aims to address the challenges of cross-domain learning of HPE without access to source data during the adaptation process. We further propose a novel framework that consists of three models: source model, intermediate model, and target model, which explores the task from both source-protect and target-relevant perspectives. The source-protect module preserves source information more effectively while resisting noise, and the target-relevant module reduces the sparsity of spatial representations by building a novel spatial probability space, and pose-specific contrastive learning and information maximization are proposed on the basis of this space. Comprehensive experiments on several domain adaptive HPE benchmarks show that the proposed method outperforms existing approaches by a considerable margin. The codes are available at https://***/davidpengucf/SFDAHPE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unveiling anomalies: transformative insights from transformer-based autoencoder models

引用

International Journal of computers and Applications 2025年第1期47卷 29-44页

作者： Dehkordi, Sajad Balali Nasri, Saeed Dami, Sina Department of Electrical Engineering Najaf Abad Branch Islamic Azad University Najaf Abad Iran Digital Processing and Machine Vision Research Center Najaf Abad Branch Islamic Azad University Najaf Abad Iran Department of Computer Engineering West Tehran Branch Islamic Azad University Tehran Iran

Nowadays, Intrusion Detection Systems (IDS) play a critical role in safeguarding networks by identifying and mitigating unauthorized access and malicious activities, yet face challenges in accurately detecting sophisticated threats while minimizing false positives. In this paper, we present a comprehensive analysis of a Transformer-based Autoencoder model for network intrusion detection using the NSL-KDD dataset. We evaluate the model's anomaly detection performance based on key metrics such as Mean Squared Error (MSE), Precision-Recall Area Under Curve (AUC), and Root Mean Squared Error (RMSE) on the test set. Our findings reveal promising results, with the model achieving low MSE, high AUC, and relatively low RMSE values, indicating its effectiveness in detecting anomalies. Additionally, we provide an in-depth examination of the model architecture, highlighting the role of encoder and decoder layers, dropout regularization, and optimization techniques such as Adam optimizer. Our analysis sheds light on the efficacy of Transformer-based Autoencoder models for network intrusion detection tasks, offering insights into their architectural design and performance evaluation. The novelty of this study lies in the implementation of a Transformer-based Autoencoder architecture for network intrusion detection, showcasing its superior performance in anomaly detection compared to traditional methods. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Med-Tuning: A New Parameter-Efficient Tuning Framework for Medical Volumetric Segmentation 7

Med-Tuning: A New Parameter-Efficient Tuning Framework for M...

引用

7th International Conference on Medical Imaging with Deep Learning, MIDL 2024

作者： Shen, Jiachen Wang, Wenxuan Chen, Chen Jiao, Jianbo Liu, Jing Zhang, Yan Song, Shanshan Li, Jiangyun School of Automation and Electrical Engineering University of Science and Technology Beijing China Center for Research in Computer Vision University of Central Florida United States School of Computer Science University of Birmingham United Kingdom Institute of Automation Chinese Academy of Sciences China

The "pre-training then fine-tuning (FT)" paradigm is widely adopted to boost the model performance of deep learning-based methods for medical volumetric segmentation. However, conventional full FT incurs high computational and memory costs. Thus, it is of increasing importance to fine-tune pre-trained models for medical volumetric segmentation tasks in a both effective and parameter-efficient manner. In this paper, we introduce a new framework named Med-Tuning to realize parameter-efficient tuning (PET) for medical volumetric segmentation task and an efficient plug-and-play module named Med-Adapter for task-specific feature extraction. With a small number of tuned parameters, our framework enhances the 2D baselines’s precision on segmentation tasks, which are pre-trained on natural images. Extensive experiments on three benchmark datasets (CT and MRI modalities) show that our method achieves better results than previous PET methods on volumetric segmentation tasks. Compared to full FT, Med-Tuning reduces the fine-tuned model parameters by up to 4×, with even better segmentation performance. Our project webpage is at https://***/Med-Tuning/. © 2024 CC-BY 4.0, J. Shen, W. Wang, C. Chen, J. Jiao, J. Liu, Y. Zhang, S. Song & J. Li.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

TopNet: Transformer-Based Object Placement Network for Image Compositing

TopNet: Transformer-Based Object Placement Network for Image...

引用

Conference on computer vision and Pattern Recognition (CVPR)

作者： Sijie Zhu Zhe Lin Scott Cohen Jason Kuen Zhifei Zhang Chen Chen Center for Research in Computer Vision University of Central Florida Adobe Research

We investigate the problem of automatically placing an object into a background image for image compositing. Given a background image and a segmented object, the goal is to train a model to predict plausible placements (location and scale) of the object for compositing. The quality of the composite image highly depends on the predicted location/scale. Existing works either generate candidate bounding boxes or apply sliding-window search using global representations from background and object images, which fail to model local information in background images. However, local clues in background images are important to determine the compatibility of placing the objects with certain locations/scales. In this paper, we propose to learn the correlation between object features and all local background features with a transformer module so that detailed information can be provided on all possible location/scale configurations. A sparse contrastive loss is further proposed to train our model with sparse supervision. Our new formulation generates a 3D heatmap indicating the plausibility of all location/scale combinations in one network forward pass, which is > 10 x faster than the previous sliding-window method. It also supports interactive search when users provide a pre-defined location or scale. The proposed method can be trained with explicit annotation or in a self-supervised manner using an off-the-shelf inpainting model, and it outperforms state-of-the-art methods significantly. User study shows that the trained model generalizes well to real-world images with diverse challenging scenes and object categories.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learned Reverse ISP with Soft Supervision 17th

Learned Reverse ISP with Soft Supervision

引用

17th European Conference on computer vision, ECCV 2022

作者： Zou, Beiji Zhang, Yue School of Computer Science and Engineering Central South University Changsha410083 China Hunan Engineering Research Center of Machine Vision and Intelligent Medicine Changsha410083 China

ISBN: (纸本)9783031250620

RAW image serves as the foundation for camera imaging, which resides at the very beginning of the pipeline that generates sRGB images. Unfortunately, owing to special considerations, the information-rich RAW images are forfeited by default in most existing applications. To regain the RAW image, some works attempt to restore RAW images from RGB images. They focus on designing handcrafted model-based methods or complicated networks, however, ignoring the special property of RAW image, i.e., high dynamic range. To make up for this deficiency, we introduce a novel soft supervision, derived from the high dynamic range. Specifically, we propose to soften the original ground-truth as a multivariate Gaussian distribution so that networks could learn much more information. Then, we introduce a soft supervision driven network (SSDNet), based on convolution and transformer, for effectively restoring RAW images from RGB images. Quantitative and qualitative results show the promising restoration performance of RGB-to-RAW. In particular, our method achieved fifth place in the S7 track of AIM Reversed ISP Challenge. The source code will be available at https://***/yuezhang98/Learned-Reverse-ISP-with-Soft-Supervision. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Sync from the Sea: Retrieving Alignable Videos from Large-Scale Datasets

arXiv

引用

arXiv 2024年

作者： Dave, Ishan Rajendrakumar Heilbron, Fabian Caba Shah, Mubarak Jenni, Simon Center for Research in Computer Vision University of Central Florida United States Adobe Research United States

Temporal video alignment aims to synchronize the key events like object interactions or action phase transitions in two videos. Such methods could benefit various video editing, processing, and understanding tasks. However, existing approaches operate under the restrictive assumption that a suitable video pair for alignment is given, significantly limiting their broader applicability. To address this, we re-pose temporal alignment as a search problem and introduce the task of Alignable Video Retrieval (AVR). Given a query video, our approach can identify well-alignable videos from a large collection of clips and temporally synchronize them to the query. To achieve this, we make three key contributions: 1) we introduce DRAQ, a video alignability indicator to identify and re-rank the best alignable video from a set of candidates;2) we propose an effective and generalizable frame-level video feature design to improve the alignment performance of several off-the-shelf feature representations, and 3) we propose a novel benchmark and evaluation protocol for AVR using cycle-consistency metrics. Our experiments on 3 datasets, including large-scale Kinetics700, demonstrate the effectiveness of our approach in identifying alignable video pairs from diverse datasets. © 2024, CC BY-NC-SA.

关键词： Video analysis

来源：评论

学校读者我要写书评

暂无评论

Sparse Points to Dense Clouds: Enhancing 3D Detection with Limited LiDAR Data

Sparse Points to Dense Clouds: Enhancing 3D Detection with L...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Aakash Kumar Chen Chen Ajmal Mian Neils Lobo Mubarak Shah Center for Research in Computer Vision University of Central Florida USA University of Western Australia

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

3D detection is a critical task that enables machines to identify and locate objects in three-dimensional space. It has a broad range of applications in several fields, including autonomous driving, robotics and augmented reality. Monocular 3D detection is attractive as it requires only a single camera, however, it lacks the accuracy and robustness required for real world applications. High resolution LiDAR on the other hand, can be expensive and lead to interference problems in heavy traffic given their active transmissions. We propose a balanced approach that combines the advantages of monocular and point cloud-based 3D detection. Our method requires only a small number of 3D points, that can be obtained from a low-cost, low-resolution sensor. Specifically, we use only 512 points, which is just 1% of a full LiDAR frame in the KITTI dataset. Our method reconstructs a complete 3D point cloud from this limited 3D information combined with a single image. The reconstructed 3D point cloud and corresponding image can be used by any multi-modal off-the-shelf detector for 3D object detection. By using the proposed network architecture with an off-the-shelf multi-modal 3D detector, the accuracy of 3D detection improves by 20% compared to the state-of-theart monocular detection methods and 6% to 9% compare to the baseline multi-modal methods on KITTI and JackRabbot datasets.

关键词： Point cloud compression Cloud computing Three-dimensional displays Laser radar Accuracy Robot vision systems Detectors Robustness Object recognition Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：