Evaluating Novel Mask-RCNN Architectures for Ear Mask Segmentation
This work addresses ear segmentation for biometric applications, but it is incremental as it builds on existing Mask-RCNN methods with minor architectural variations.
The paper tackled the problem of ear segmentation for biometric recognition by comparing three newer Mask-RCNN architectures against the state-of-the-art model, finding that the newer models outperformed it in Average Precision scores but no single model performed best across multiple datasets.
The human ear is generally universal, collectible, distinct, and permanent. Ear-based biometric recognition is a niche and recent approach that is being explored. For any ear-based biometric algorithm to perform well, ear detection and segmentation need to be accurately performed. While significant work has been done in existing literature for bounding boxes, a lack of approaches output a segmentation mask for ears. This paper trains and compares three newer models to the state-of-the-art MaskRCNN (ResNet 101 +FPN) model across four different datasets. The Average Precision (AP) scores reported show that the newer models outperform the state-of-the-art but no one model performs the best over multiple datasets.