Hybrid Task Cascade for Instance Segmentation

TLDR

Cascade is a classic yet powerful architecture that has boosted performance on various tasks. The authors propose the Hybrid Task Cascade (HTC) framework to integrate detection and segmentation in a joint multi‑stage process, addressing the open question of how to apply cascade to instance segmentation. HTC interweaves detection and segmentation across stages, using a fully convolutional branch for spatial context and progressively learning discriminative features while integrating complementary ones at each stage. HTC improves mask AP by 38.4% and 1.5% over a strong Cascade Mask R‑CNN baseline on MSCOCO, achieving 48.6 AP and ranking first in the COCO 2018 Challenge. Code is available at https://github.com/open-mmlab/mmdetection.

Abstract

Cascade is a classic yet powerful architecture that has boosted performance on various tasks. However, how to introduce cascade to instance segmentation remains an open question. A simple combination of Cascade R-CNN and Mask R-CNN only brings limited gain. In exploring a more effective approach, we find that the key to a successful instance segmentation cascade is to fully leverage the reciprocal relationship between detection and segmentation. In this work, we propose a new framework, Hybrid Task Cascade (HTC), which differs in two important aspects: (1) instead of performing cascaded refinement on these two tasks separately, it interweaves them for a joint multi-stage processing; (2) it adopts a fully convolutional branch to provide spatial context, which can help distinguishing hard foreground from cluttered background. Overall, this framework can learn more discriminative features progressively while integrating complementary features together in each stage. Without bells and whistles, a single HTC obtains 38.4% and 1.5% improvement over a strong Cascade Mask R-CNN baseline on MSCOCO dataset. Moreover, our overall system achieves 48.6 mask AP on the test-challenge split, ranking 1st in the COCO 2018 Challenge Object Detection Task. Code is available at https://github.com/open-mmlab/mmdetection.