大ゼミスライド作り

コードを下のgitのものに書き換え中

GitHub - YutaroOgawa/pytorch_advanced: 書籍「つくりながら学ぶ! PyTorchによる発展ディープラーニング」の実装コードを配置したリポジトリです

１．Try to use wavelet transform time-frequency diagram, S-transform time-frequency diagram, W-V transform time-frequency diagram and short-time Fourier time-frequency diagram as data sets for deep learning.

2.Try to improve CNN-LSTM network parameters and reduce training time.

3.Begin to write the paper framework of ICIAE.

4. … Continue Reading ››

1.I am writing my AROB thesis.

2.I am running the DCGAN program to try to complete the image generation.

Read three reviews this week.

Single-View 3D reconstruction: A Survey of deep learning methods

A Review of 3D Reconstruction Methods for Visual Deep Learning
Review of 3D reconstruction algorithm research

A Review of 3D Reconstruction Methods for Visual Deep Learning

I am writing a paper recently, and I felt a little confused about expressing the skills I used in this paper. So I read some papers like "Deep layer aggregation", "Center3D", "Objects as points", "CornerNet" and "Deformable Convolutional Networks". The paper is still under writing.

Uniformer（https://openreview.net/forum?id=nBU_u6DLvoK）:

Introduction:

Combining the ability of aggregation feature context of 3D convolution with the self-attention mechanism of visual transformer, to improve the ability of remote dependence of the model. Uniformer combines the advantages of both and achieves good results in many visual tasks.