1

Unsupervised domain adaptation in semantic segmentation alleviates the reliance on expensive pixel-wise annotation. It uses a labeled …

Finding accurate correspondences among different views is the Achilles heel of unsupervised Multi-View Stereo (MVS). Existing methods …

The success of monocular 3D object detection highly relies on considerable labeled data, which is costly to obtain. To alleviate the …

In this work, we study the continual semantic segmentation problem, where the deep neural networks are required to incorporate new …

Person re-identification (ReID) has gained an impressiveprogress in recent years. However, the occlusion is still acommon and …

In this paper, we present a novel approach to synthesize realistic images based on their semantic layouts. The core idea is that for …

Real-world data often exhibit imbalanced distributions, where certain target values have significantly fewer observations. Existing …

Traditional convolution-based generative adversarial networks synthesize images based on hierarchical local operations, where …

Unpaired image-to-image translation (I2I) has achieved great success in various applications. However, its generalization capacity is …

Unpaired image-to-image translation aims to translate images from the source class to target one by providing sufficient data for these …