output
20062026
most citedDistance-IoU Loss: Faster and Better Learning for Bounding Box Regression

961 citations

Showing 2019 · cs.CVShow all

15 papers · 2 filters

cs.CV2019★ 3 cited

Exploiting Operation Importance for Differentiable Neural Architecture Search

Xukai Xie, Yuan Zhou, Sun-Yuan Kung

Recently, differentiable neural architecture search methods significantly reduce the search cost by constructing a super network and relax the architecture representation by assign…

cs.CV2019★ 961 cited

Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression

Zhaohui Zheng, Ping Wang, Wei Liu +3

Bounding box regression is the crucial step in object detection. In existing methods, while -norm loss is widely adopted for bounding box regression, it is not tailored to…

cs.CV2019★ 62 cited

Semi-Heterogeneous Three-Way Joint Embedding Network for Sketch-Based Image Retrieval

Jianjun Lei, Yuxin Song, Bo Peng +3

Sketch-based image retrieval (SBIR) is a challenging task due to the large cross-domain gap between sketches and natural images. How to align abstract sketches and natural images i…

cs.CV2019

Comb Convolution for Efficient Convolutional Architecture

Dandan Li, Yuan Zhou, Shuwei Huo +1

Convolutional neural networks (CNNs) are inherently suffering from massively redundant computation (FLOPs) due to the dense connection pattern between feature maps and convolution…

cs.CV2019★ 16 cited

Facial Expression Restoration Based on Improved Graph Convolutional Networks

Zhilei Liu, Le Li, Yunpeng Wu +1

Facial expression analysis in the wild is challenging when the facial image is with low resolution or partial occlusion. Considering the correlations among different facial local r…

cs.CV2019★ 87 cited

Relation Modeling with Graph Convolutional Networks for Facial Action Unit Detection

Zhilei Liu, Jiahui Dong, Cuicui Zhang +2

Most existing AU detection works considering AU relationships are relying on probabilistic graphical models with manually extracted features. This paper proposes an end-to-end deep…