Jeffkang-94 / faster-rcnn-bdd100k

Object detection bdd100k with faster-rcnn

faster-rcnn-bdd100k

Object detection bdd100k with faster-rcnn

Prerequisites

• Pytorch >= 1.1
• torchvision >= 0.5

directory structure :

-| BDD100k
   -| images
     -| 100k
       -| train
       -| test
       -| val
   -| labels
.......

Setting up Config

By default, all paths and hyperparameters are loaded from cfg.py. bdd_path is a data path, which might need to be changed depending on the user profile. idx=0 means, bdd100k datasets will be used during the training.

##########  User specific settings ##########################
bdd_path = "/mnt2/datasets/bdd100k" # datapath


batch_size = 64

num_epochs = 25
lr = 0.001
ckpt = False
model_name = "bdd100k_24.pth"
##############################################################

idx = 0
dset_list = ["bdd100k", "Cityscapes"]
ds = dset_list[idx]

Start to train

get the datalist

if idx=0, you can get the bdd dataset list under the datalist folder.

python get_datalist.py
-| datalists
  -| bdd100k_train_images_path.txt
  -| bdd100k_val_images_path.txt

Specify the data transform

You can modify the transform composition.
Refer to bdd.py under datasets folder.

def get_transform(train):
    transforms = []
    transforms.append(T.ToTensor())
    transforms.append(T.resize((256,512)))
    transforms.append(T.RandomHorizontalFlip(0.5))
    
    return T.Compose(transforms)

Training

Support for baseline has been added. Domain adaptive features will be added later.

python train_baseline.py

Note

Note that, you have to change the min, max value of size if user wants to use resize transform.
Default option provided by torchvision, has min_size=800, max_size=1333.

def get_model(num_classes):
    #model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)
    model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True, min_size=256, max_size=512, image_mean=[0.5,0.5,0.5], image_std=[0.5,0.5,0.5])
    in_features = model.roi_heads.box_predictor.cls_score.in_features
    model.roi_heads.box_predictor = torchvision.models.detection.faster_rcnn.FastRCNNPredictor(
        in_features, num_classes
    )  # replace the pre-trained head with a new one
    return model.cuda()

Evaluation

Evaluation in performed in COCO format. Users need to specify saved model_name in cfg.pyon which evaluation is supposed to occur.

CocoAPI needs to be compiled. first download it from here

$ cd cocoapi/PythonAPI
$ python setup.py build_ext install

Now evaluation can be performed.

$ python3 evaluation_script.py

Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.148 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.286 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.129 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.031 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.175 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.370 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.117 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.208 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.219 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.071 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.276 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.480

Pre-trained model

Not supported, yet.

Example

Not supported, yet.