I finished the representation of AROB last week, and I found many excellent representation of other doctors and professors. I found that I have to practice more to make my representation to a perfect one in the future.

I learned about the docker, and tried a some small applications with docker, I find that docker is a powerful software like gitlab we used before.

I finished the powerpoint for AROB, and prepared for the representation on Thursday this week.

I read the official documents of pytorch for several days. I read the meaning of tensor in pytorch which is similar to numpy in python. and There is a good connection between pytorch and python. Later, I need to learned some basic knowledge of docker.

I modified my AROB thesis, and submit the thesis to the AROB last week. From the 3D object detection survey, I learned a lot about the 3D detection works, and find the frameworks using cameras would be better than those using lidars.

Next I would try … Continue Reading ››

	

先週の進捗（LI ZHEN）

I tried the program of "3D detection using deep hough voting". At first I used the trained dataset of SUNRGBD to run the program. I get some results datas and images.

先週の進捗(李鎮)

I did some conclusion on 3D detection methods, and learn a lot from the survey thesis:Deep Learning for 3D Point Clouds: A Survey

I tried to run pointnet program on my ubunu system. But a problem always occurred many times. ubuntu-desktop always broke many times.

The abstract of my thesis in AROB2021 is completed

I learned about PointNet and PointNet++ from 3D detection thesis "Deep Hough voting for 3d object detection in point clouds". I continued read the thesis "PointNet++:Deep Hierarchical Feature Learning on Point Sets in a Metric Space". This is the basic work … Continue Reading ››

I finished the abstract of the thesis "3D Object Detection on Intelligent Driving: A Survey".

I learned some Japanese lesson on the web by my self.

I tryed to running SA-SSA program in my virtual environment, and I encountered a little problem about the spconv1.0, there is always some problem while compiling spconv1.0. There could be some relationships between spconv1.0 and my cudnn version. I would solve this problem soon.
I am also downloading some datasets form KITTI.

3D検出アルゴリズムの研究

日本語を勉強します