CODEWITHSUNDEEP
conv-neural-networkobject-detection
Hadi GhahremanNezhad
Hadi GhahremanNezhad

Reputation: 2455

is Faster RCNN rotation invariant?

I want to generate a dataset of vehicles to train in networks such as RCNN, Fast RCNN, Faster RCNN, etc. My question is:

Does the angle of vehicles in the original image matter for the detector? I mean if the dataset only contains vehicle images which are all directed from left to right, can the trained network detect any vehicle with any angle?

Upvotes: 1

Views: 1260

Answers (3)

Maryam Bahrami
Maryam Bahrami

Reputation: 1104

Detectron2 added Rotated Faster RCNN network recently. To create such model, you should create annotations for vehicles with rotated bounding box, which is:

rbbox = [center_x, center_x, width, height, angle]

Visit this link for more information.

Upvotes: 1

Hadi GhahremanNezhad
Hadi GhahremanNezhad

Reputation: 2455

The answer to this question is no. faster rcnn has only one option for augmentation which is side to side flipping. You have to add rotation to it by yourself. You can train the network with a lot of samples of rotated objects with different angles.

Upvotes: 1

alexander-liu7
alexander-liu7

Reputation: 201

No！ Because CNN's rotation invariance is limited, Faster RCNN which is based on CNN can't detect object whose orientaion is too different from the orientation of examples in training set.

Upvotes: 2

Related Questions