Stage: Mainly engaged in low-quality multi-scale image quantization convolution and MIPS architecture chip algorithm design， based on RSIC-V architecture.The main task of this project is to improve the combination of research and engineering through academic guidance on the notes of systematic problems in experimental engineering, and through the core course of QAS in-depth learning.
GMM https://petewarden.com/2015/04/20/why-gemm-is-at-the-heart-of-deep-learning/
Opencv https://docs.opencv.org/3.4.5/d9/df8/tutorial_root.html
Opencv http://www.opencv.org.cn/opencvdoc/2.3.2/html/doc/tutorials/tutorials.html
object detection：http://machinethink.net/blog/object-detection/
（Conference and Workshop on Neural Information Processing Systems）NIPS ：https://papers.nips.cc/book/advances-in-neural-information-processing-systems-31-2018
ICLR ：https://openreview.net/group?id=ICLR.cc/2019/Conference CVPR ：http://cvpr2019.thecvf.com/ ICCV ：http://iccv2019.thecvf.com/ ECCV ：https://eccv2018.org/
machinelearningbook：https://datawhalechina.github.io/pumpkin-book/#/
https://github.com/naturomics/CapsNet-Tensorflow https://github.com/bojone/Capsule
https://github.com/Dikea/ML-Weekly-Learning/tree/master/CV
https://github.com/Eric3911/Code-with-Life
https://github.com/Eric3911/Coding-learning
https://github.com/MarkMoHR/Awesome-Edge-Detection-Papers
https://arxiv.org/list/cs.AI/recent https://github.com/Eric3911/Paper-Sharing https://github.com/Eric3911/Super-Paper https://github.com/Eric3911/Engineering-papers
One of the future directions of deep learning: Based on the combination of nonlinear dynamics and fractional nonlinear equations in group theory, a meta learning method, model safe federated learning and data safe blockchain are proposed after reinforcement learning.
https://github.com/Eric3911/CDCS
https://github.com/Eric3911/Computer_vision_Object
https://github.com/Eric3911/Visual-Science-Competition
https://mmcheng.net/tag/salient-object-detection/
Jim Blinn's Corner: Dirty Pixels
Jim Blinn's Corner: A Trip Down The Graphics Pipeline
Jim Blinn's Corner: Notation, Notation, Notation
Millimeter wave radar fusion of optical flow field and ray tracing in 77GHz phase control matrix
It has the most engineering significance and academic value to use fan / attention recognition after using super-resolution and antagonism to generate Song typeface for character recognition, and the FOTS model based on reinforcement learning made by Mr. Jin is also a good direction.
https://github.com/chongyangtao/Awesome-Scene-Text-Recognition https://github.com/Eric3911/icpr2018_ocr_papers
https://github.com/DmitryUlyanov/deep-image-prior
Open coursera: https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/