Rain Xu

Interested to be a hacker, right now is learning Web Application and Web Security

STN-Spatial Transformer Network(Image Classification) and Deformable Convolution Networks

STN https://arxiv.org/abs/1506.02025 from 2015. STN helps to crop out and scale-normalizes the appropriate region, which can simplify the subsequent classfication task and lead to better classific...

Posted by Rain on April 15, 2020

Curriculum learning

https://ronan.collobert.com/pub/matos/2009_curriculum_icml.pdf Curriculum learning describes a type of learning in which you first start out with only easy examples of a task and then gradually in...

Posted by Rain on April 15, 2020

Multi-task Learning

Multi-task learning https://ruder.io/multi-task/ ## hard parameter sharing It is generally applied by sharing the hidden layer between all tasks, while keeping several task-specific output layers...

Posted by Rain on April 15, 2020

landmark detection literature survey

landmark detection

paper: 2018: https://arxiv.org/abs/1805.05563 2019 RetinaFace: Single-stage Dense Face Localisation in the Wild. 2017 RetinaNet: https://towardsdatascience.com/review-retinanet-focal-loss-obj...

Posted by Rain on April 15, 2020

Over Fitting and Regularization and Optimization

Regularization it basically adds the penalty as model complexity increases. Regularization parameter(lambda) penalizes all the parameters except intercept so that model generalizes the data and w...

Posted by Rain on April 15, 2020

Git notes

resource from https://linuxhint.com/git-shallow-clone-and-clone-depth/#:~:text=Git’s%20solution%20to%20the%20problem,lot%20of%20space%20and%20time. Shallow Clone: where you can use clone depth to ...

Posted by Rain on April 15, 2020

Small detector model comparison

2015: single shot detector(SSD) YOLO: you only look once ### Fast Yolo 9 of 24 CNN layers from GoogleNet fully connected detector network real-time systems on PASCAL VOC 2007 2016: ...

Posted by Rain on April 15, 2020

Yolo

1st, YOLO us extrmely fast. Since we frame detection as a regression problem we don’t need a complex pipeline. We simply run our neural network on a new image at test time to predict detections....

Posted by Rain on April 15, 2020

Object Detection in Pytorch

Objection Detection Tutorial in Pytorch

source from https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Object-Detection The model that are going to be implemented is SSD(single shot Multibox Detector) Original Implementation:https://g...

Posted by Rain on April 15, 2020

Node_js_notes

Mobilenet network

weird things happens, when I first run a node js server script in my VM, it works good, then I let my computer to sleep mode. And seconds later I realized that might kill the node js process and I ...

Posted by Rain on April 15, 2020

ABOUT ME

A computer science student

✉️ sddy@gmail.com