Feifan Lv

I am an algorithm engineer at Huawei, where I work on computer vision and autonomous driving.

Before that, I received my Master's degree from Beihang University, under the supervision of Prof. Feng Lu. I was a research intern in Microsoft Research Asia during Summer and Autumn, 2019.

Email  /  Google Scholar  /  Github

I'm interested in image processing and computer vision. To be specific, I'm particularly interested in low-level vision tasks including denoising, enhancement and super-resolution.

By introducing two attention maps to guide the exposure enhancement and denoising tasks respectively, our enhancement network can work in an input adaptive way.

A light-weight CNN is proposed for non-uniform illumination image enhancement, which can handle color, exposure, contrast, noise and artifacts, etc., simultaneously and effectively.

To produces clear color images in 24 hours, we separate the VIS and NIR information from mixed signals, and enhance the VIS signal adaptively with the NIR signal as assistance.

To approximate the response of the InGaAs sensor by exploiting the largely ignored sensitivity of a Silicon sensor, weak as it is, in the SWIR range.

Inspired by Koch's Postulates, we define a pathological descriptor and propose a GAN based method to visualize the symptom and feature encoded in this descriptor.

To achieve fast and accurate perception, we propose an efficient CNN called Multiply Spatial Fusion Network (MSFNet) using Class Boundary Supervision.

Enhancing low-light image/video using the proposed multi-branch network by extracting and fusing rich features up to different levels.

Last update: Dec. 2021  |  Template stolen from Jon Barron