• YouTube

Description

I talk about the "How to achieve the 3D object recognition accuracy 80%(40 category) for 3month "

Deep Learning is the good technique for image recognition and speech recognition. And it apply the other field. Many people try to apply the Deep Learning, but it is difficult to make a result.

In my situation, I have enough knowledge about the 3D object and label data. I'll talk about the how to achieve the 80 % (40 category)

In My approach

1. Getting the Information
   • How to choose the information
   • How to choose framework
2. Getting the Data
   • Public data
   • How to make the own data
3. Try small
   • Trying the small data set
   • Trying the train and predict
4. Deciding the direction focus
   • Choose what you can control
5. Prioritizing with high certainty
   • Pre-process
   • Improve the train speed
6. Increasing the challenge times
   • Using the GPU
   • CPU optimization
   • multi process
   • resource
7. Parameter Tuning
   • Improve Model Versatility or Improve Data Versatility
   • Model Tuning
     • RandomDropOut
     • LeakyRelu
   • Data Argumantion
8. Product
   • Minimum function
   • Using Docker

I hope to people who want to apply Deep Learning for the 3D model