Faster Training of Mask R-CNN by Focusing on Instance Boundaries

Abstract

We present an auxiliary task to Mask R-CNN, an instance segmentation network, which leads to faster training of the mask head. Our addition to Mask R-CNN is a new prediction head, the Edge Agreement Head, which is inspired by the way human annotators perform instance segmentation. Human annotators copy the contour of an object instance and only indirectly the occupied instance area. Hence, the edges of instance masks are particularly useful as they characterize the instance well. The Edge Agreement Head therefore encourages predicted masks to have similar image gradients to the ground-truth mask using edge detection filters. We provide a detailed survey of loss combinations and show improvements on the MS COCO Mask metrics compared to using no additional loss. Our approach marginally increases the model size and adds no additional trainable model variables. While the computational costs are increased slightly, the increment is negligible considering the high computational cost of the Mask R-CNN architecture. As the additional network head is only relevant during training, inference speed remains unchanged compared to Mask R-CNN. In a default Mask R-CNN setup, we achieve a training speed-up and a relative overall improvement of 8.1% on the MS COCO metrics compared to the baseline.

arXiv comment9 pages, 7 figures, 5 tables

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Fully Convolutional Networks for Semantic Segmentation
• An Overview of Multi-Task Learning in Deep Neural Networks
• Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
• GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks
• Facial Landmark Detection: a Literature Survey
• Holistically-Nested Edge Detection
• Auxiliary Tasks in Multi-task Learning
• Object Contour Detection with a Fully Convolutional Encoder-Decoder Network
• InstanceCut: from Edges to Instances with MultiCut

Cited by in corpus (8)

• Deblending and Classifying Astronomical Sources with Mask R-CNN Deep Learning
• CFUN: Combining Faster R-CNN and U-net Network for Efficient Whole Heart Segmentation
• Boundary-preserving Mask R-CNN
• The Devil is in the Boundary: Exploiting Boundary Representation for Basis-based Instance Segmentation
• Multi Receptive Field Network for Semantic Segmentation
• Learning Crisp Edge Detector Using Logical Refinement Network
• SE-PSNet: Silhouette-based Enhancement Feature for Panoptic Segmentation Network
• A Turbo-Inference Strategy for Object Detection and Instance Segmentation