L
Linchao Zhu
Researcher at University of Technology, Sydney
Publications - 75
Citations - 3740
Linchao Zhu is an academic researcher from University of Technology, Sydney. The author has contributed to research in topics: Object detection & Feature (computer vision). The author has an hindex of 21, co-authored 75 publications receiving 1704 citations. Previous affiliations of Linchao Zhu include Australian Artificial Intelligence Institute & Google.
Papers
More filters
Proceedings ArticleDOI
ActBERT: Learning Global-Local Video-Text Representations
Linchao Zhu,Yi Yang +1 more
TL;DR: This paper introduces ActBERT for self-supervised learning of joint video-text representations from unlabeled data and introduces an ENtangled Transformer block to encode three sources of information, i.e., global actions, local regional objects, and linguistic descriptions.
Proceedings ArticleDOI
Entangled Transformer for Image Captioning
TL;DR: A Transformer-based sequence modeling framework built only with attention layers and feedforward layers that enables the Transformer to exploit semantic and visual information simultaneously and achieves state-of-the-art performance on the MSCOCO image captioning dataset.
Proceedings ArticleDOI
Auto-ReID: Searching for a Part-Aware ConvNet for Person Re-Identification
TL;DR: This work proposes a retrieval-based search algorithm over a specifically designed reID search space, named Auto-ReID, which enables the automated approach to find an efficient and effective CNN architecture for reID.
Journal ArticleDOI
Uncovering the Temporal Context for Video Question Answering
TL;DR: An encoder–decoder approach using Recurrent Neural Networks to learn the temporal structures of videos and introduce a dual-channel ranking loss to answer multiple-choice questions is presented.
Proceedings ArticleDOI
Learning Filter Pruning Criteria for Deep Convolutional Neural Networks Acceleration
TL;DR: Learning Filter Pruning Criteria (LFPC) is proposed, which develops a differentiable pruning criteria sampler that is learnable and optimized by the validation loss of the pruned network obtained from the sampled criteria.