We propose a flow-guided transformer, which innovatively leverage the motion discrepancy
exposed by optical flows to instruct the attention retrieval in transformer for high fidelity video …

Multi-viewpoint image stitching aims to stitch images taken from different viewpoints into
pictures with a broader field of view. The stitched images are subject to artifacts, geometric …

Physical objects have inertia, which resists changes in the velocity and motion direction.
Inspired by this, we introduce inertia prior that optical flow, which reflects object motion in a …

Bridging global context interactions correctly is important for high-fidelity image completion
with large masks. Previous methods attempting this via deep or large receptive field (RF) …

Sketch-based image manipulation is an interactive image editing task to modify an image
based on input sketches from users. Existing methods typically convert this task into a …

With the rapid growth of in-the-wild videos taken by non-specialists, blind video quality
assessment (VQA) has become a challenging and demanding problem. Although lots of …

Image inpainting aims to fill the missing hole of the input. It is hard to solve this task
efficiently when facing high-resolution images due to two reasons:(1) Large reception field …

Image inpainting seeks a semantically consistent way to recover the corrupted image in the
light of its unmasked content. Previous approaches usually reuse the well-trained GAN as …

Scene understanding in adverse weather conditions (eg, rainy and foggy days) has drawn
increasing attention, arising some specific benchmarks and algorithms. However, scene …

Night-Time Scene Parsing (NTSP) is essential to many vision applications, especially for
autonomous driving. Most of the existing methods are proposed for day-time scene parsing …