Fine grained identification, classification, retrieval data set collation

Copyright : This paper is written by 【 Mohism AI】 Original launch 、 Dear readers 、 Please refer to 、 Thank you Sanlian
Statement : As the whole network AI field One of the bloggers with the most dry goods ,️ Live up to your time ️

The main data sets used in fine-grained correlation research are as follows

Fine-grained visual classification (FGVC) The relevant main training and evaluation data sets are as follows ：

• CUB-200-2011
• Standford Cars
• FGVC-Aircraft

Weak supervision In fine-grained classification task , The sample dataset structure is as follows

Here the data set structure ： Need to combine papers 、 Code , Reorganize the downloaded dataset images and labels

dataset
├── train
│   ├── class_001
|   |      ├── 1.jpg
|   |      ├── 2.jpg
|   |      └── ...
│   ├── class_002
|   |      ├── 1.jpg
|   |      ├── 2.jpg
|   |      └── ...
│   └── ...
└── test
    ├── class_001
    |      ├── 1.jpg
    |      ├── 2.jpg
    |      └── ...
    ├── class_002
    |      ├── 1.jpg
    |      ├── 2.jpg
    |      └── ...
    └── ...

The download links for the three datasets are as follows

---

• https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/
• http://www.vision.caltech.edu/datasets/
• http://ai.stanford.edu/~jkrause/cars/car_dataset.html

FGVC-Aircraft download

---

• https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/

CUB-200-2011 download

---

Data set introduction

Caltech-UCSD Birds-200-2011 (CUB-200-2011) is an extended version of the CUB-200 dataset, with roughly double the number of images per class and new part location annotations.

• Number of categories: 200
• Number of images: 11,788
• Annotations per image: 15 Part Locations, 312 Binary Attributes, 1 Bounding Box
• For detailed information about the dataset, please see the technical report linked below.

download

• http://www.vision.caltech.edu/datasets/

Standford Cars download

---

• http://ai.stanford.edu/~jkrause/cars/car_dataset.html

---

Select columns

---

• Deep learning model training reasoning —— Basic environment construction, recommended blog review order 【 Foundation installation — I've carefully sorted it out for everyone 】——【1024 Special issue 】

In the field of computer vision Eight columns 、 A lot of dry goods 、 If you are interested, you can learn about

• ️ Image style conversion —— Code environment construction Practical course 【 Follow and read 】！
• Image restoration - Code environment construction - Knowledge summary Practical course 【 It's said to be ok 】
• Supersession reconstruction - Code environment construction - Knowledge summary How to make the white moonlight clearer 【 Single artifact 】
• YOLO special column , Only actual combat , be unreasonable Image classification 【 Recommended collection 】！

• Deep learning ： Environment building , Article to read

• Deep learning ： Interesting learning, deep learning

• Deploy applications on the ground ： Transformation of model deployment - Speed up - encapsulation

• CV and Voice data sets ： Data set collation

• give the thumbs-up Collection Leaving a message. Bloggers insist on writing 、 The biggest motivation to update high-quality blog posts ！