Although zero-shot learning (ZSL) has gained widespread concern due to its excellent capacity of recognizing new object classes without seeing any visual instances, most existing methods assume that all seen-class instances used for training are correctly labeled. In some real application scenarios, ...
Webly Supervised Learning Meets Zero-shot Learning: A Hybrid Approach for Fine-grained Classification Li Niu, Ashok Veeraraghavan, and Ashu Sabharwal Department of Electrical and Computer Engineering, Rice University {ln7,vashok,ashu}@rice.edu Abstract Fine-grained image classification, which targets ...
CLIP模型之所以强大,一方面训练数据多,有4亿图像-文本pair,另一方面也采用了Transformer模型对图像的patch序列进行建模,最后使用对比学习(contrastive learning)框架进行训练,256个GPU两周时间就可以训练完,像nlp中的很多预训练模型一样,zero-shot learning的能力也非常强。从目前的一些demo看出,DALLE的效果十分惊艳,当然也...
Less is more: zero-shot learning from online textual documents with noise suppression∗ Ruizhi Qiao, Lingqiao Liu, Chunhua Shen, Anton van den Hengel School of Computer Science, The University of Adelaide, Australia {ruizhi.qiao,lingqiao.liu,chunhua.shen,anton.vandenhengel}@adelaide.edu.au ...
Further zero-shot learning experiments demonstrate the benefits and limitations of this kind of approaches in the challenging setting of data scarcity and noisy labels for the set of seen classes. This chapter combines and extends our ongoing work on NoisyArt dataset.Chiaro, Riccardo Del...
ML-SKG:Chung-Wei Lee, Wei Fang, Chih-Kuan Yeh, Yu-Chiang Frank Wang. "Multi-Label Zero-Shot Learning With Structured Knowledge Graphs." CVPR (2018). [pdf] [project] GZSL-SE:Vinay Kumar Verma, Gundeep Arora, Ashish Mishra, Piyush Rai. "Generalized Zero-Shot Learning via Synthesized Examp...
correlations between input features and prediction labels. Then, self-pruning contrastive learning is...
In zero-shot learning (ZSL) community, it is generally recognized that transductive learning performs better than inductive one as the unseen-class samples are also used in its training stage. How to generate pseudo labels for unseen-class samples and how to use such usually noisy pseudo labels...
There are increasing efforts to address the problem of insufficient or even no labeled instances, such as one-shot and few-shot learning [23] deal with the classes of few labeled instances; open world recognition performs the tasks: detecting the novelty of the test classes via open set ...
Zero-shot Learning Using Multimodal Descriptions Utkarsh Mall, Bharath Hariharan, Kavita Bala Cornell University {utkarshm, bharathh, kb}@cs.cornell.edu Abstract Zero-shot learning (ZSL) tackles the problem of recog- nizing unseen classes using only semantic descriptions, e.g., attributes...