Zhen-Liang Ni

TL;DR: In this paper, an attention-guided network is proposed to segment the cataract surgical instrument, which captures global context and encodes semantic dependencies to emphasize key semantic features, boosting the feature representation.

TL;DR: A novel network, Refined Attention Segmentation Network, is proposed to simultaneously segment surgical instruments and identify their categories and the U-shape network which is popular in segmentation is used.

TL;DR: The results confirm the initial hypothesis, namely that algorithm performance degrades with an increasing domain gap, and suggest that future research should concentrate on detection and segmentation of small, crossing, moving and transparent instrument(s) (parts).

TL;DR: An attention-guided lightweight network (LWANet), which can segment surgical instruments in real-time while takes little computational costs is proposed.

TL;DR: A novel network, Pyramid Attention Aggregation Network, is proposed to aggregate multi-scale attentive features for surgical instruments, which learns the shape and size features of surgical instruments in different receptive fields and thus addresses the scale variation issue.