Trajectory-aware Principal Manifold Framework for Data Augmentation and Image Generation

no code implementations • 30 Jul 2023 • Elvis Han Cui, Bingbin Li, Yanan Li, Weng Kee Wong, Donghui Wang

Many existing methods generate new samples from a parametric distribution, like the Gaussian, with little attention to generate samples along the data manifold in either the input or feature space.

Data Augmentation Image Generation

Dual Path Structural Contrastive Embeddings for Learning Novel Objects

no code implementations • 23 Dec 2021 • Bingbin Li, Elvis Han Cui, Yanan Li, Donghui Wang, Weng Kee Wong

Learning novel classes from a very few labeled samples has attracted increasing attention in machine learning areas.

Meta-Learning Transfer Learning

Class-Incremental Few-Shot Object Detection

no code implementations • 17 May 2021 • Pengyang Li, Yanan Li, Han Cui, Donghui Wang

To tackle this problem, we propose a novel method LEAST, which can transfer with Less forgetting, fEwer training resources, And Stronger Transfer capability.

Clustering Few-Shot Object Detection +3

Fine-grained Semantic Constraint in Image Synthesis

no code implementations • 12 Jan 2021 • Pengyang Li, Donghui Wang

With mask as prior, the model in this paper is constrained so that the generated images conform to visual senses, which will reduce the unexpected diversity of samples generated from the generative adversarial network.

Attribute Generative Adversarial Network +1

MGD-GAN: Text-to-Pedestrian generation through Multi-Grained Discrimination

no code implementations • 2 Oct 2020 • Shengyu Zhang, Donghui Wang, Zhou Zhao, Siliang Tang, Di Xie, Fei Wu

In this paper, we investigate the problem of text-to-pedestrian synthesis, which has many potential applications in art, design, and video surveillance.

Generative Adversarial Network Image Generation

Zero-Shot Learning with Generative Latent Prototype Model

no code implementations • 26 May 2017 • Yanan Li, Donghui Wang

Zero-shot learning, which studies the problem of object classification for categories for which we have no training examples, is gaining increasing attention from community.

Transfer Learning Zero-Shot Learning

Learning Robust Features with Incremental Auto-Encoders

no code implementations • 26 May 2017 • Yanan Li, Donghui Wang

In this paper, we propose a new method to learn non-linear robust features by taking advantage of the data manifold structure.

Zero-Shot Recognition using Dual Visual-Semantic Mapping Paths

no code implementations • CVPR 2017 • Yanan Li, Donghui Wang, Huanhang Hu, Yuetan Lin, Yueting Zhuang

This mapping is learned on training data of seen classes and is expected to have transfer ability to unseen classes.

Zero-Shot Learning

Task-driven Visual Saliency and Attention-based Visual Question Answering

no code implementations • 22 Feb 2017 • Yuetan Lin, Zhangyang Pang, Donghui Wang, Yueting Zhuang

Visual question answering (VQA) has witnessed great progress since May, 2015 as a classic problem unifying visual and textual data into a system.

Question Answering Visual Question Answering