院の授業を受けた

自己分析を少し整理した

NTT西日本→合格

富士通の最終面接が4/20にあるので、それで就職活動終了します。

・オービック3次面接
・ntt西日本　最終面接

予定
・オービック社員座談会
・アドソル社員座談会

研究したいです。。。

I completed the JIIAE journal, and plan to submit it.

I read the google's fusion object detection work, and read the 3D object detection part:Objectron: A Large Scale Dataset of Object-Centric Videos in the Wild with Pose Annotations and MobilePose: Real-Time Pose Estimation for Unseen Objects with Weak … Continue Reading ››

1.Read paper: 视觉问答中的多模态信息融合_庞章阳.pdf

注意力机制海洋场景图像理解算法_邬满.pdf

融合空间注意力机制的图像语义描述算法_郭列.pdf

基于多模态注意力机制的图像理解描述新方法_李学明.pdf

2.recurrent "pytorch-tutorial-image-caption "

3.make a demo on RNN、 LSTM

I finished the AROB journal, and plan to submit it this week.

I finished the IIAE journal, and plan to submit it soon.

1.read paper：UNITER:UNIVERSAL IMAGE-TEXT REPRESENT LEARNING

https://mountain.elcs.kyutech.ac.jp/gao/read-paper/-/blob/master/UNITER_UNiversal_Image-TExt.pdf

2.read paper：

Show and Tell: Lessons learned from the 2015
MSCOCO Image Captioning Challenge

3.recurrent “2”

1. It took nearly a week to move, buy and tidy up
2. read paper： Ans wer_Questions_with_Right_Image_Regions_A_Visual.pdf

3.learn the code of VL-BERT

I finally modified the AROB journal, and still not submit it. And the journal of IIAE is still not finished.

I practiced the representation on 3.23 by myself.

I learned some image processing skills before the neural network.

Last week, I followed a latest CVPR paper: Track to Detect and Segment: An Online Multi-Object Tracker. I followed this work, and realized it. This work concentrates tracking, detecting, segmenting together, and can also detect a 3D model of objects. Besides, all the work can fluency run on videos.