Skip to content Skip to navigation

Qualifying Exam: Zero-shot Learning from Noisy Text Descriptions


The main shortcoming of deep learning methods is the inevitable requirement of large-scale labeled training data that need to be collected and annotated by costly human labor. Zero-shot learning aims to recognize objects with no training samples available by introducing semantic representations of classes. We propose two methods to tackle zero-shot learning problem from noisy text descriptions as semantic representations of classes. The first one is a visual-semantic embedding-based method that embeds visual data and semantic data to a shared embedding space, while connecting text terms to its relevant visual parts. The second one is a GAN(Generative Adversarial Network)-based method that is able to generate visual feature based on semantic information, and the synthetic visual features are then used as training samples for unseen class classification.

Yizhe Zhu
Event Date: 
01/31/2019 - 12:00pm
Prof. Ahmed Elgammal (chair), Prof. Gerard de Melo, Prof. Yongfeng Zhang, Prof. Badri Nath
Event Type: 
Qualifying Exam
Dept. of Computer Science