Keras implementation of RetinaNet object detection as described in Focal Loss for Dense Object Detection by Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He and Piotr Dollár.

1. Clone this repository.
2. In the repository, execute python setup.py install --user. Note that due to inconsistencies with how tensorflow should be installed, this package does not define a dependency on tensorflow as it will try to install that (which at least on Arch linux results in an incorrect installation). Please make sure tensorflow is installed as per your systems requirements. Also, make sure Keras 2.0.9 is installed.
3. As of writing, this repository requires the master branch of keras-resnet (run pip install --user --upgrade git+https://github.com/broadinstitute/keras-resnet).
4. Optionally, install pycocotools if you want to train / test on the MS COCO dataset. Clone the cocoapi repository and inside the PythonAPI folder, execute python setup.py install --user.

An example on how to train keras-retinanet can be found here.

For training on Pascal VOC, run:

python examples/train_pascal.py <path to VOCdevkit/VOC2007>

For training on MS COCO, run:

python examples/train_coco.py <path to MS COCO>

For training on a custom dataset, a CSV file can be used as a way to pass the data. To train using your CSV, run:

python examples/train_csv.py <path to csv file containing annotations> <path to csv file containing classes>

The expected format of each line of the annotations CSV is:

filepath,x1,y1,x2,y2,class_name

Images with multiple bounding boxes should use one row per bounding box. For example:

/data/imgs/img_001.jpg,837,346,981,456,cow
/data/imgs/img_002.jpg,215,312,279,391,cat
/data/imgs/img_002.jpg,22,5,89,84,bird

Note that indexing for pixel values starts at 0. The expected format of each line of the classes CSV is:

class_name,id

Indexing for classes starts at 0. Do not include a background class as it is implicit. For example:

cow,0
cat,1
bird,2

In general, the steps to train on your own datasets are:

1. Create a model by calling for instance keras_retinanet.models.ResNet50RetinaNet and compile it. Empirically, the following compile arguments have been found to work well:

model.compile(
    loss={
        'regression'    : keras_retinanet.losses.regression_loss,
        'classification': keras_retinanet.losses.focal_loss()
    },
    optimizer=keras.optimizers.adam(lr=1e-5, clipnorm=0.001)
)

2. Create generators for training and testing data (an example is show in keras_retinanet.preprocessing.PascalVocGenerator).
3. Use model.fit_generator to start training.

An example of testing the network can be seen in this Notebook. In general, output can be retrieved from the network as follows:

_, _, detections = model.predict_on_batch(inputs)

Where detections are the resulting detections, shaped (None, None, 4 + num_classes) (for (x1, y1, x2, y2, cls1, cls2, ...)).

Loading models can be done in the following manner:

from keras_retinanet.models.resnet import custom_objects
model = keras.models.load_model('/path/to/model.h5', custom_objects=custom_objects)

Execution time on NVIDIA Pascal Titan X is roughly 55msec for an image of shape 1000x600x3.

The MS COCO model can be downloaded here. Results using the cocoapi are shown below (note: according to the paper, this configuration should achieve a mAP of 0.343).

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.325
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.513
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.342
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.149
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.354
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.465
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.288
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.437
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.464
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.263
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.510
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.623

• The examples show how to train keras-retinanet on Pascal VOC and MS COCO. Example output images are shown below.

• Configure CI

• This repository requires Keras 2.0.9.
• This repository is tested using OpenCV 3.3 (3.0+ should be supported).

Contributions to this project are welcome.

Feel free to join the #keras-retinanet Keras Slack channel for discussions and questions.