Li’s Computer Vision Blogs

Difference between Different Neural Networks

Autoencoder is a simple 3-layer neural network where output units are directly connected back to input units. E.g. in a network like this:

How to read research papers

A summary from this video:

Tools for Researching

As a research, learning to track papers and keeping ourselves updated is definitely necessary. Tens of papers needed to be scanned every week, so these tools and find a comfortable way to do is worth the time to figure out. For me, there are three main steps: searching, reading and…

Unsupervised segmentation tools

1. Mean-shift based segmentation:

Companies that Computer Vision related Engineers can work for

Today I was make arrangement to attend CVPR 2018 conference hold in Salt Lake city, Utah, USA, and the following companies are from the exhibitor list on this top conference. I think it is a pretty complete list for the…

GANs, cGANs, AE, AAE, AVE, CAAE, CAVE

The difference between each other.

Computer Vision Application

All categories are benefiting from deep learning. And deep learning is continuingly creating beyond of…

Graph-based Computer Vision Algorithm

In computer vision, an image is usually modeled as a graph wherein each pixel or superpixel is a vertex and each vertex is connected to other defined neighbors (the most commonly is 4, up, down, left, right, there is also fully connected). Based on different…