G
Guosheng Lin
Researcher at Nanyang Technological University
Publications - 193
Citations - 12611
Guosheng Lin is an academic researcher from Nanyang Technological University. The author has contributed to research in topics: Computer science & Segmentation. The author has an hindex of 36, co-authored 153 publications receiving 8618 citations. Previous affiliations of Guosheng Lin include Salesforce.com & Association for Computing Machinery.
Papers
More filters
Book ChapterDOI
IntegratedPIFu: Integrated Pixel Aligned Implicit Function for Single-View Human Reconstruction
TL;DR: IntegratedPIFu as mentioned in this paper proposes depth-oriented sampling, a novel training scheme that improves any pixel-aligned implicit model's ability to reconstruct important human features without noisy artefacts, and it is able to improve the structural correctness of reconstructed meshes.
Journal ArticleDOI
Depth and Video Segmentation Based Visual Attention for Embodied Question Answering
TL;DR: This work proposes a depth and segmentation based visual attention mechanism for Embodied Question Answering that effectively boosts the performance of the VQA module and navigation module, leading to 4.9% and 5.6% overall improvement in EQA accuracy on House3D and Matterport3D datasets respectively.
Book ChapterDOI
Dynamically Transformed Instance Normalization Network for Generalizable Person Re-Identification
Bingliang Jiao,Ling-Hong Liu,Liying Gao,Guosheng Lin,Lu Yang,Shizhou Zhang,Peng Wang,Yanning Zhang +7 more
TL;DR: In this article , the authors proposed a new normalization scheme called Dynamically Transformed Instance Normalization (DTIN), which employs dynamic convolution to allow the unnormalized feature to control the transformation of the normalized features into new representations.
Journal ArticleDOI
Feature flow: In-network feature flow estimation for video object detection
TL;DR: In this article, the authors propose a novel network (IFF-Net) with an In-network Feature Flow estimation module (IFF module) for video object detection, which is able to directly produce feature flow which indicates the feature displacement.
Dissertation
Structured output prediction and binary code learning in computer vision.
TL;DR: This paper aims to demonstrate the efforts towards in-situ applicability of EMMARM, which aims to provide real-time information about the physical properties of the response of the immune system to attacks.