7,497 Hits in 5.7 sec

Attribute-Based Synthetic Network (ABS-Net): Learning more from pseudo feature representations

Jiang Lu, Jin Li, Ziang Yan, Fenghua Mei, Changshui Zhang
2018 Pattern Recognition  
In this paper we present a simple but high-performance ZSL approach by generating pseudo feature representations (GPFR).  ...  Zero-shot learning (ZSL) is a challenging task aiming at recognizing novel classes without any training instances.  ...  Pseudo feature representations generation.  ... 
doi:10.1016/j.patcog.2018.03.006 fatcat:zqo2pjvwsfgmjjrkn5dsrccfya

Harnessing Object and Scene Semantics for Large-Scale Video Understanding

Zuxuan Wu, Yanwei Fu, Yu-Gang Jiang, Leonid Sigal
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We illustrate effectiveness of this semantic representation through experiments on zero-shot action/video classification and clustering.  ...  Our semantic fusion network combines three streams of information using a three-layer neural network: (i) frame-based low-level CNN features, (ii) object features from a state-of-the-art large-scale CNN  ...  Zero-shot Learning Baselines: We compare the following methods for largescale zero-shot recognition: 1. DAP-word.  ... 
doi:10.1109/cvpr.2016.339 dblp:conf/cvpr/WuFJS16 fatcat:6zpkaj64c5hx7g6hsttamlj5yq

Non-generative Generalized Zero-shot Learning via Task-correlated Disentanglement and Controllable Samples Synthesis [article]

Yaogong Feng, Xiaowen Huang, Pengbo Yang, Jian Yu, Jitao Sang
2022 arXiv   pre-print
Synthesizing pseudo samples is currently the most effective way to solve the Generalized Zero-Shot Learning (GZSL) problem.  ...  ones by adversarial learning of domain adaption towards reasonable synthesis; (2) Controllable pseudo sample synthesis, to synthesize edge-pseudo and center-pseudo samples with certain characteristics  ...  Acknowledgments This work is supported by the National Key R&D Program of China (Grant No. 2018AAA0100604), the National Natural Science Foundation of China (Grant No. 61832004, 61632002), and Beijing  ... 
arXiv:2203.05335v3 fatcat:w74w2jzp4nh53nmwgye2efkbim

Unifying Unsupervised Domain Adaptation and Zero-Shot Visual Recognition [article]

Qian Wang, Penghui Bu, Toby P. Breckon
2019 arXiv   pre-print
This is a generalized zero-shot learning problem where the side information comes from the source domain in the form of labelled samples instead of class-level semantic representations commonly used in  ...  traditional zero-shot learning.  ...  Zero-Shot Learning Zero-shot learning (ZSL) aims to recognize novel classes by transferring knowledge learned from known classes to unseen classes [17] .  ... 
arXiv:1903.10601v2 fatcat:d35y4p4tgrh5tcvrbwwrn6euwu

A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts [article]

Yizhe Zhu, Mohamed Elhoseiny, Bingchen Liu, Xi Peng, Ahmed Elgammal
2018 arXiv   pre-print
With added pseudo data, zero-shot learning is naturally converted to a traditional classification problem.  ...  Most existing zero-shot learning methods consider the problem as a visual semantic embedding one.  ...  Generalized Zero-Shot Learning The conventional zero-shot recognition considers that queries come from only unseen classes.  ... 
arXiv:1712.01381v3 fatcat:q4ykcflujbfqtbe5w3udfs7eny

Progressive Ensemble Networks for Zero-Shot Recognition [article]

Meng Ye, Yuhong Guo
2019 arXiv   pre-print
It can naturally bridge the domain shift problem in visual appearances and be extended to the generalized zero-shot learning scenario.  ...  The ensemble network is built by learning multiple image classification functions with a shared feature extraction network but different label embedding representations, which enhance the diversity of  ...  Transductive Zero-Shot Learning Different from the standard zero-shot learning setting where unlabeled instances from unseen classes are treated as inaccessible in the training phase, transductive ZSL  ... 
arXiv:1805.07473v2 fatcat:otex6hdwt5ftpmqu2nixdp25zm

SIGN: Spatial-information Incorporated Generative Network for Generalized Zero-shot Semantic Segmentation [article]

Jiaxin Cheng, Soumyaroop Nandi, Prem Natarajan, Wael Abd-Almageed
2021 arXiv   pre-print
Furthermore, while self-training is widely used in zero-shot semantic segmentation to generate pseudo-labels, we propose a new knowledge-distillation-inspired self-training strategy, namely Annealed Self-Training  ...  Unlike conventional zero-shot classification, zero-shot semantic segmentation predicts a class label at the pixel level instead of the image level.  ...  Acknowledgement This material is based on research sponsored by Air Force Research Laboratory (AFRL) under agreement number FA8750-19-1-1000.  ... 
arXiv:2108.12517v1 fatcat:ui7no7xnxvhuxfzdrv7l7iq2w4

Few-Shot Adaptation for Multimedia Semantic Indexing [article]

Nakamasa Inoue, Koichi Shinoda
2018 arXiv   pre-print
Few-shot adaptation provides robust parameter estimation with few training examples, by optimizing the parameters of zero-shot learning and supervised many-shot learning simultaneously.  ...  In this method, first we build a zero-shot detector, and then update it by using the few examples.  ...  We focus on how to generate pseudo training samples that give a zero-shot detector as a result of supervised learning.  ... 
arXiv:1807.07203v1 fatcat:xn2e3v44kvcalaub2ogpkwrkki

Few-Shot Adaptation for Multimedia Semantic Indexing

Nakamasa Inoue, Koichi Shinoda
2018 2018 ACM Multimedia Conference on Multimedia Conference - MM '18  
Few-shot adaptation provides robust parameter estimation with few training examples, by optimizing the parameters of zero-shot learning and supervised manyshot learning simultaneously.  ...  We propose a few-shot adaptation framework, which bridges zeroshot learning and supervised many-shot learning, for semantic indexing of image and video data.  ...  We focus on how to generate pseudo training samples that give a zero-shot detector as a result of supervised learning.  ... 
doi:10.1145/3240508.3240592 dblp:conf/mm/InoueS18 fatcat:c2f7sik6cfegnc66ihbvw4bloe

Learn from Anywhere: Rethinking Generalized Zero-Shot Learning with Limited Supervision [article]

Gaurav Bhatt, Shivam Chandhok, Vineeth N Balasubramanian
2021 arXiv   pre-print
to improve generalization in any-shot learning.  ...  A common problem with most zero and few-shot learning approaches is they suffer from bias towards seen classes resulting in sub-optimal performance.  ...  In addition to results on generalized zero-shot learning (GZSL) and few-shot learning, our use of AUD allows us to demonstrate the applicability of our proposed method for generalized zero-shot in a limited  ... 
arXiv:2107.04952v2 fatcat:nydifnfnfbgybegvky55rfkoyy

DenseCLIP: Extract Free Dense Labels from CLIP [article]

Chong Zhou, Chen Change Loy, Bo Dai
2021 arXiv   pre-print
By adding pseudo labeling and self-training, DenseCLIP+ surpasses SOTA transductive zero-shot semantic segmentation methods by large margins, e.g., mIoUs of unseen classes on PASCAL VOC/PASCAL Context/  ...  Contrastive Language-Image Pre-training (CLIP) has made a remarkable breakthrough in open-vocabulary zero-shot image recognition.  ...  Zero-Shot Visual Recognition. Zero-shot learning aims at classifying instances of those categories that are not seen during training.  ... 
arXiv:2112.01071v1 fatcat:4kzhdstgqngrpgsntxb7aqtbca

Multimodal Clustering Networks for Self-supervised Learning from Unlabeled Videos [article]

Brian Chen, Andrew Rouditchenko, Kevin Duarte, Hilde Kuehne, Samuel Thomas, Angie Boggust, Rameswar Panda, Brian Kingsbury, Rogerio Feris, David Harwath, James Glass, Michael Picheny (+1 others)
2021 arXiv   pre-print
To evaluate our approach, we train our model on the HowTo100M dataset and evaluate its zero-shot retrieval capabilities in two challenging domains, namely text-to-video retrieval, and temporal action localization  ...  In this context, this paper proposes a self-supervised training framework that learns a common multimodal embedding space that, in addition to sharing representations across different modalities, enforces  ...  This work is supported by IARPA via DOI/IBC contract number D17PC00341.  ... 
arXiv:2104.12671v3 fatcat:3sgcrya54ndrto3xabpwszw3ra

Unsupervised Image Classification for Deep Representation Learning [article]

Weijie Chen and Shiliang Pu and Di Xie and Shicai Yang and Yilu Guo and Luojun Lin
2020 arXiv   pre-print
Deep clustering against self-supervised learning is a very important and promising direction for unsupervised visual representation learning since it requires little domain knowledge to design pretext  ...  and few-shot image classification.  ...  Iteratively alternating Eq.4 and Eq.2 for pseudo label generation and representation learning, can it really learn a disentangled representation?  ... 
arXiv:2006.11480v2 fatcat:y73sgvo7k5cl7aixdeaduwydfa

RegionCLIP: Region-based Language-Image Pretraining [article]

Yiwu Zhong, Jianwei Yang, Pengchuan Zhang, Chunyuan Li, Noel Codella, Liunian Harold Li, Luowei Zhou, Xiyang Dai, Lu Yuan, Yin Li, Jianfeng Gao
2021 arXiv   pre-print
Moreoever, the learned region representations support zero-shot inference for object detection, showing promising results on both COCO and LVIS datasets.  ...  Contrastive language-image pretraining (CLIP) using image-text pairs has achieved impressive results on image classification in both zero-shot and transfer learning settings.  ...  Different from OVR, we propose to learn visual region representation from our "pseudo" region-text pairs given by another pretrained CLIP model.  ... 
arXiv:2112.09106v1 fatcat:3pypzvqrhnhodkn5iz26qbguj4

A Generative Approach to Zero-Shot and Few-Shot Action Recognition [article]

Ashish Mishra, Vinay Kumar Verma, M Shiva Krishna Reddy, Arulkumar S, Piyush Rai, Anurag Mittal
2018 arXiv   pre-print
) and generalized zero-shot learning settings.  ...  We present a generative framework for zero-shot action recognition where some of the possible action classes do not occur in the training data.  ...  This approach helps to reduce the baisness in the case of Generalize Zero-Shot Learning.  ... 
arXiv:1801.09086v1 fatcat:m5pdvxu57fd47or3lraqdlnsze
« Previous Showing results 1 — 15 out of 7,497 results