Výsledky vyhledávání - IEEE/CVF Conference on Computer Vision AND Pattern Recognition

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018) : Salt Lake City, Utah, USA 18-22 June 2018 Autor IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers

Získat plný text

Collecting Cross-Modal Presence-Absence Evidence for Weakly-Supervised Audio- Visual Event Perception Autor Gao, Junyu, Chen, Mengyuan, Xu, Changsheng

“…With only video-level event labels, this paper targets at the task of weakly-supervised audio-visual event perception (WS-AVEP), which aims to temporally…”
Získat plný text

Masked-attention Mask Transformer for Universal Image Segmentation Autor Cheng, Bowen, Misra, Ishan, Schwing, Alexander G., Kirillov, Alexander, Girdhar, Rohit

“… Each choice of semantics defines a task. While only the semantics of each task differ, current research focuses on designing spe-cialized architectures for each task…”
Získat plný text

Bottleneck Transformers for Visual Recognition Autor Srinivas, Aravind, Lin, Tsung-Yi, Parmar, Niki, Shlens, Jonathon, Abbeel, Pieter, Vaswani, Ashish

“…We present BoTNet, a conceptually simple yet powerful backbone architecture that incorporates self-attention for multiple computer vision tasks including image classification, object detection and instance segmentation…”
Získat plný text

GLIGEN: Open-Set Grounded Text-to-Image Generation Autor Li, Yuheng, Liu, Haotian, Wu, Qingyang, Mu, Fangzhou, Yang, Jianwei, Gao, Jianfeng, Li, Chunyuan, Lee, Yong Jae

“… enabling them to also be conditioned on grounding inputs. To preserve the vast concept knowledge of the pre-trained model, we freeze all of its weights and inject the grounding information into new trainable layers via a gated mechanism…”
Získat plný text

Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detection Autor Haliassos, Alexandros, Vougioukas, Konstantinos, Petridis, Stavros, Pantic, Maja

“… Some recent works show improvements in generalisation but rely on cues that are easily corrupted by common post-processing operations such as compression…”
Získat plný text

Dense Contrastive Learning for Self-Supervised Visual Pre-Training Autor Wang, Xinlong, Zhang, Rufeng, Shen, Chunhua, Kong, Tao, Li, Lei

“…To date, most existing self-supervised learning methods are designed and optimized for image classification. These pre-trained models can be sub-optimal for…”
Získat plný text

Deformable ConvNets V2: More Deformable, Better Results Autor Zhu, Xizhou, Hu, Han, Lin, Stephen, Dai, Jifeng

“… To address this problem, we present a reformulation of Deformable ConvNets that improves its ability to focus on pertinent image regions, through increased modeling power and stronger training…”
Získat plný text

Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks Autor Dong, Yinpeng, Pang, Tianyu, Su, Hang, Zhu, Jun

“…Deep neural networks are vulnerable to adversarial examples, which can mislead classifiers by adding imperceptible perturbations. An intriguing property of…”
Získat plný text

Google Landmarks Dataset v2 - A Large-Scale Benchmark for Instance-Level Recognition and Retrieval Autor Weyand, Tobias, Araujo, Andre, Cao, Bingyi, Sim, Jack

“…While image retrieval and instance recognition techniques are progressing rapidly, there is a need for challenging datasets to accurately measure their performance -- while posing novel challenges…”
Získat plný text

Deformable Siamese Attention Networks for Visual Object Tracking Autor Yu, Yuechen, Xiong, Yilei, Huang, Weilin, Scott, Matthew R.

“…Siamese-based trackers have achieved excellent performance on visual object tracking…”
Získat plný text

MobileOne: An Improved One millisecond Mobile Backbone Autor Vasu, Pavan Kumar Anasosalu, Gabriel, James, Zhu, Jeff, Tuzel, Oncel, Ranjan, Anurag

“… However, these metrics may not correlate well with latency of the network when deployed on a mobile device…”
Získat plný text

Multi-class Token Transformer for Weakly Supervised Semantic Segmentation Autor Xu, Lian, Ouyang, Wanli, Bennamoun, Mohammed, Boussaid, Farid, Xu, Dan

“…). Inspired by the fact that the attended regions of the one-class token in the standard vision transformer can be leveraged to form a class-agnostic localization map, we investigate if the transformer…”
Získat plný text

TopFormer: Token Pyramid Transformer for Mobile Semantic Segmentation Autor Zhang, Wenqiang, Huang, Zilong, Luo, Guozhong, Chen, Tao, Wang, Xinggang, Liu, Wenyu, Yu, Gang, Shen, Chunhua

“…Although vision transformers (ViTs) have achieved great success in computer vision, the heavy computational cost hampers their applications to dense prediction tasks such as semantic segmentation on mobile devices…”
Získat plný text

Causality Inspired Representation Learning for Domain Generalization Autor Lv, Fangrui, Liang, Jian, Li, Shuang, Zang, Bin, Liu, Chi Harold, Wang, Ziteng, Liu, Di

“…Domain generalization (DG) is essentially an out-of-distribution problem, aiming to generalize the knowledge learned from multiple source domains to an unseen…”
Získat plný text

Deep Snake for Real-Time Instance Segmentation Autor Peng, Sida, Jiang, Wen, Pi, Huaijin, Li, Xiuli, Bao, Hujun, Zhou, Xiaowei

“… For structured feature learning on the contour, we propose to use circular convolution in deep snake, which better exploits the cycle-graph structure of a contour compared against generic graph convolution…”
Získat plný text

Token Contrast for Weakly-Supervised Semantic Segmentation Autor Ru, Lixiang, Zheng, Heliang, Zhan, Yibing, Du, Bo

“… Though the recent Vision Transformer (ViT) can remedy this flaw, we observe it also brings the over-smoothing issue, i.e…”
Získat plný text

Finding Task-Relevant Features for Few-Shot Learning by Category Traversal Autor Li, Hongyang, Eigen, David, Dodge, Samuel, Zeiler, Matthew, Wang, Xiaogang

“…Few-shot learning is an important area of research. Conceptually, humans are readily able to understand new concepts given just a few examples, while in more…”
Získat plný text

Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction Autor Ma, Tiezheng, Nie, Yongwei, Long, Chengjiang, Zhang, Qing, Li, Guiqing

“… Our method is based on the observation that a good "initial guess" of the future poses is very helpful in improving the forecasting accuracy…”
Získat plný text

Wavelet Integrated CNNs for Noise-Robust Image Classification Autor Li, Qiufu, Shen, Linlin, Guo, Sheng, Lai, Zhihui

“…Convolutional Neural Networks (CNNs) are generally prone to noise interruptions, i.e., small image noise can cause drastic changes in the output. To suppress…”
Získat plný text