Výsledky vyhľadávania - Deep learning architectures and techniques; Recognition: detection

Equalized Focal Loss for Dense Long-Tailed Object Detection Autor Li, Bo, Yao, Yongqiang, Tan, Jingru, Zhang, Gang, Yu, Fengwei, Lu, Jianwei, Luo, Ye

“…Despite the recent success of long-tailed object detection, almost all long-tailed object detectors are developed based on the two-stage paradigm…”
Získať plný text

Sylph: A Hypernetwork Framework for Incremental Few-shot Object Detection Autor Yin, Li, Perez-Rua, Juan M, Liang, Kevin J

“…We study the challenging incremental few-shot object de-tection (iFSD) setting. Recently, hypernetwork-based approaches have been studied in the context of…”
Získať plný text

A ConvNet for the 2020s Autor Liu, Zhuang, Mao, Hanzi, Wu, Chao-Yuan, Feichtenhofer, Christoph, Darrell, Trevor, Xie, Saining

“…The "Roaring 20s" of visual recognition began with the introduction of Vision Transformers (ViTs…”
Získať plný text

Grounded Language-Image Pre-training Autor Li, Liunian Harold, Zhang, Pengchuan, Zhang, Haotian, Yang, Jianwei, Li, Chunyuan, Zhong, Yiwu, Wang, Lijuan, Yuan, Lu, Zhang, Lei, Hwang, Jenq-Neng, Chang, Kai-Wei, Gao, Jianfeng

“…This paper presents a grounded language-image pretraining (GLIP) model for learning object-level, language-aware, and semantic-rich visual representations…”
Získať plný text

CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows Autor Dong, Xiaoyi, Bao, Jianmin, Chen, Dongdong, Zhang, Weiming, Yu, Nenghai, Yuan, Lu, Chen, Dong, Guo, Baining

“…We present CSWin Transformer, an efficient and effective Transformer-based backbone for general-purpose vision tasks. A challenging issue in Transformer design…”
Získať plný text

MetaFormer is Actually What You Need for Vision Autor Yu, Weihao, Luo, Mi, Zhou, Pan, Si, Chenyang, Zhou, Yichen, Wang, Xinchao, Feng, Jiashi, Yan, Shuicheng

“… Based on this observation, we hypothesize that the general architecture of the transformers, instead of the specific token mixer module, is more essential to the model's performance…”
Získať plný text

beta-DARTS: Beta-Decay Regularization for Differentiable Architecture Search Autor Ye, Peng, Li, Baopu, Li, Yikang, Chen, Tao, Fan, Jiayuan, Ouyang, Wanli

“…Neural Architecture Search (NAS) has attracted increasingly more attention in recent years because of its capability to design deep neural network automatically…”
Získať plný text

SLIC: Self-Supervised Learning with Iterative Clustering for Human Action Videos Autor Khorasgani, Salar Hosseini, Chen, Yuxuan, Shkurti, Florian

“…Self-supervised methods have significantly closed the gap with end-to-end supervised learning for image classification [13], [24…”
Získať plný text

Revisiting Weakly Supervised Pre-Training of Visual Perception Models Autor Singh, Mannat, Gustafson, Laura, Adcock, Aaron, De Freitas Reis, Vinicius, Gedik, Bugra, Kosaraju, Raj Prateek, Mahajan, Dhruv, Girshick, Ross, Dollar, Piotr, Van Der Maaten, Laurens

“…Model pre-training is a cornerstone of modern visual recognition systems. Although fully supervised pre-training on datasets like ImageNet is still the de-facto…”
Získať plný text

Multimodal Token Fusion for Vision Transformers Autor Wang, Yikai, Chen, Xinghao, Cao, Lele, Huang, Wenbing, Sun, Fuchun, Wang, Yunhe

“…Many adaptations of transformers have emerged to address the single-modal vision tasks, where self-attention modules are stacked to handle input sources like…”
Získať plný text

Knowledge Distillation via the Target-aware Transformer Autor Lin, Sihao, Xie, Hongwei, Wang, Bing, Yu, Kaicheng, Chang, Xiaojun, Liang, Xiaodan, Wang, Gang

“… However, people tend to overlook the fact that, due to the architecture differences, the semantic information on the same spatial location usually vary…”
Získať plný text

Single-Domain Generalized Object Detection in Urban Scene via Cyclic-Disentangled Self-Distillation Autor Wu, Aming, Deng, Cheng

“… And we consider a realistic yet challenging scenario, namely Single-Domain Generalized Object Detection (Single-DGOD…”
Získať plný text

TransMix: Attend to Mix for Vision Transformers Autor Chen, Jie-Neng, Sun, Shuyang, He, Ju, Torr, Philip, Yuille, Alan, Bai, Song

“…Mixup-based augmentation has been found to be effective for generalizing models during training, especially for Vision Transformers (ViTs) since they can…”
Získať plný text

Unbiased Teacher v2: Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors Autor Liu, Yen-Cheng, Ma, Chih-Yao, Kira, Zsolt

“…With the recent development of Semi-Supervised Object Detection (SS-OD) techniques, object detectors can be improved by using a limited amount of labeled data and abundant unlabeled data…”
Získať plný text

MiniViT: Compressing Vision Transformers with Weight Multiplexing Autor Zhang, Jinnian, Peng, Houwen, Wu, Kan, Liu, Mengchen, Xiao, Bin, Fu, Jianlong, Yuan, Lu

“…Vision Transformer (ViT) models have recently drawn much attention in computer vision due to their high model capability. However, ViT models suffer from huge…”
Získať plný text

TableFormer: Table Structure Understanding with Transformers Autor Nassar, Ahmed, Livathinos, Nikolaos, Lysak, Maksym, Staar, Peter

“… In this paper, we present a new table-structure identification model. The latter improves the latest end-to-end deep learning model (i.e…”
Získať plný text

VISTA: Boosting 3D Object Detection via Dual Cross-VIew SpaTial Attention Autor Deng, Shengheng, Liang, Zhihao, Sun, Lin, Jia, Kui

“…Detecting objects from LiDAR point clouds is of tremendous significance in autonomous driving…”
Získať plný text

Human-Object Interaction Detection via Disentangled Transformer Autor Zhou, Desen, Liu, Zhichao, Wang, Jian, Wang, Leshan, Hu, Tao, Ding, Errui, Wang, Jingdong

“…Human-Object Interaction Detection tackles the problem of joint localization and classification of human object interactions…”
Získať plný text

Progressive End-to-End Object Detection in Crowded Scenes Autor Zheng, Anlin, Zhang, Yuang, Zhang, Xiangyu, Qi, Xiaojuan, Sun, Jian

“…In this paper, we propose a new query-based detection framework for crowd detection…”
Získať plný text

DTA: Physical Camouflage Attacks using Differentiable Transformation Network Autor Suryanto, Naufal, Kim, Yongsu, Kang, Hyoeun, Larasati, Harashta Tatimma, Yun, Youngyeo, Le, Thi-Thu-Huong, Yang, Hunmin, Oh, Se-Yoon, Kim, Howon

“… In this paper, we propose the Differentiable Transformation Attack (DTA), a framework for generating a robust physical adversarial pattern on a target object to camouflage it against object detection models with a wide range of transformations…”
Získať plný text