https://github.com/joydeepmedhi/anchor-boxes-with-kmeans
How to initialize Anchors in Faster RCNN for custom dataset?
https://github.com/joydeepmedhi/anchor-boxes-with-kmeans
anchor anchor-box aspect-ratio bounding-boxes clusters computer-vision custom-dataset detection distance-metric faster-rcnn hyperparameters iou kmeans object-detection python tensorflow-models
Last synced: over 1 year ago
JSON representation
How to initialize Anchors in Faster RCNN for custom dataset?
- Host: GitHub
- URL: https://github.com/joydeepmedhi/anchor-boxes-with-kmeans
- Owner: joydeepmedhi
- Created: 2018-08-21T10:41:00.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2020-08-18T15:39:01.000Z (about 6 years ago)
- Last Synced: 2025-04-07T02:05:48.915Z (over 1 year ago)
- Topics: anchor, anchor-box, aspect-ratio, bounding-boxes, clusters, computer-vision, custom-dataset, detection, distance-metric, faster-rcnn, hyperparameters, iou, kmeans, object-detection, python, tensorflow-models
- Language: Jupyter Notebook
- Size: 5.77 MB
- Stars: 91
- Watchers: 6
- Forks: 31
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Initial Anchor Boxes Estimation using KMeans Clusterring for Faster-RCNN
[](https://www.codefactor.io/repository/github/joydeepmedhi/anchor-boxes-with-kmeans)
## Introduction
Faster-RCNN is one of the **state-of-the-art** object detection algorithms around.
If you are not familiar with Faster-RCNN, Please go through [this blog](https://tryolabs.com/blog/2018/01/18/faster-r-cnn-down-the-rabbit-hole-of-modern-object-detection/).
Here is the link to the original paper [ Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://arxiv.org/abs/1506.01497).
When we train Faster RCNN for custom datasets, we often get confused over how to choose hyperparameters for the Network. Anchor boxes (one of the hyperparameters) are very important to detect objects with different scales and aspect ratios. We will get improved detection results if we get the anchors right.
The training & hypereparameters are in accordance with **Tensorflow Object Detection [API](https://github.com/tensorflow/models/tree/master/research/object_detection)**.
### Faster-RCNN config file
```
faster_rcnn{
# other hyperparameters
first_stage_anchor_generator {
grid_anchor_generator {
height: 256
width: 256
height_stride: 16
width_stride: 16
scales: 0.9
scales: 1.14
scales: 1.53
aspect_ratios: .8
aspect_ratios: 1.15
aspect_ratios: 2.77
}
}
}
```
#### height & width
This is the size of base anchor size. (i.e. for scale 1 and aspect ratio 1, base anchor is 256 x 256)
#### height_stride & width_stride
This is basically the stride of anchor centers. Generally, we want to visit each point of the feature map (final convolutional layer) and create a set of anchors. Hence, It is the *subsampling ratio* of the network. In case of **VGG16** this ratio is 16. Different network archetectures have different subsampling ratios. User may select this stride as per the base-model or use case.
#### scales & aspect_ratios
Aspect Ratio of an anchor box is basically *width/height*. Scales are bigger as the anchor box are from base box (i.e. 512 x 512 box is twice as big as 256 x 256).
```
if aspect_ratio = ar
base_anchor = 256 x 256
"width_b x height_b" is the dimension of an anchor box
width_b = scale * sqrt(ar) * base_anchor[0]
height_b = scale * base_anchor[1] / sqrt(ar)
```
---
## Analysis of bounding boxes (Training data)
1. Convert the *XML* files to a *csv* file.
``` xml_to_csv.py``` (modify this file as per your *XML* format)
2. Open ```EDA_of_bbox.ipynb ``` **jupyter notebook** for analysis.
Here, we convert the image dimension with *_compute_new_static_size()* function. Then we normalize bounding box height and width according to new image dimension.
Then we find optimal clusters and cluster centers using **K-Means**. This is inspired from [YOLO](https://pjreddie.com/darknet/yolo/).
Distribution of Bounding Boxes!
### Experiments
#### 1
Cluster bbox (width, height) on **eucledian** distance metric
Blue Line - Base Model (cards dataset)
Red Line - Cluster Model (cards dataset)
#### 2
Cluster bbox (width, height) on **iou** metric (This is prefered as eucledian distance metric will give priority to bigger boxes and minimize their loss)
Blue Line - Base Model (cards dataset)
Pink Line - IOU Cluster Model (cards dataset)
#### 3
Cluster **AR** and **Scales** of bbox Separately with **eucledian** distance metric.
Blue Line - Base Model (cards dataset)
Green Line - Cluster Model (cards dataset)
*************** **More to be added** *****************
---
## References
1. [KMeans in YOLO](https://lars76.github.io/object-detection/k-means-anchor-boxes/)
2. [Cards Dataset (Reference)](https://github.com/EdjeElectronics/TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10)
3. [Advantage & Disadvantage of KMeans](http://playwidtech.blogspot.com/2013/02/k-means-clustering-advantages-and.html)
4. [Different Clusterring Algorithms](https://towardsdatascience.com/the-5-clustering-algorithms-data-scientists-need-to-know-a36d136ef68)