Skip to main content
Fig. 2 | EURASIP Journal on Image and Video Processing

Fig. 2

From: Anchor-free object detection with mask attention

Fig. 2

Describe the proposed method. By stacking multiple hourglass network and mask modules to enhance the network extraction ability, the mask module can get the image segmentation result. The corner detection module has two parts, and the top left and bottom right corners of the object are obtained

Back to article page