Implicit Feature Refinement for Instance Segmentation

  title={Implicit Feature Refinement for Instance Segmentation},
  author={Lufan Ma and Tiancai Wang and Bin Dong and Jiangpeng Yan and Xiu Li and Xiangyu Zhang},
  journal={Proceedings of the 29th ACM International Conference on Multimedia},
We propose a novel implicit feature refinement module for high-quality instance segmentation. Existing image/video instance segmentation methods rely on explicitly stacked convolutions to refine instance features before the final prediction. In this paper, we first give an empirical comparison of different refinement strategies, which reveals that the widely-used four consecutive convolutions are not necessary. As an alternative, weight-sharing convolution blocks provides competitive… 

UniInst: Unique Representation for End-to-End Instance Segmentation

  • Yimin OuRui Yang Xiu Li
  • Computer Science
  • 2022
This work designs an instance-aware one-to-one assignment scheme, namely OYOR, which dynamically assigns one unique representation to each instance according to the matching quality between predictions and ground truths, and achieves competitive performance against mainstream methods on COCO test-dev.



A Coarse-to-Fine Instance Segmentation Network with Learning Boundary Representation

A holistic boundary-aware branch is designed and instance-agnostic supervision is introduced to assist regression and the proposed coarse-to-fine module achieves competitive performance compared to existing boundary-based methods with a lightweight design and a simple pipeline.

The Devil is in the Boundary: Exploiting Boundary Representation for Basis-based Instance Segmentation

This work proposes Boundary Basis based Instance Segmentation (B2Inst) to learn a global boundary representation that can complement existing global-mask-based methods that are often lacking high-frequency details, and devise a unified quality measure of both mask and boundary.

Mask Encoding for Single Shot Instance Segmentation

Instead of predicting the two-dimensional mask directly, MEInst distills it into a compact and fixed-dimensional representation vector, which allows the instance segmentation task to be incorporated into one-stage bounding-box detectors and results in a simple yet efficient instance segmentations framework.

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

This work addresses the task of semantic image segmentation with Deep Learning and proposes atrousΒ spatial pyramid pooling (ASPP), which is proposed to robustly segment objects at multiple scales, and improves the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models.

SipMask: Spatial Information Preservation for Fast Image and Video Instance Segmentation

A fast single-stage instance segmentation method that preserves instance-specific spatial information by separating mask prediction of an instance to different sub-regions of a detected bounding-box, leading to improved mask predictions and a mask alignment weighting loss and a feature alignment scheme to better correlate mask prediction with object detection.

Hybrid Task Cascade for Instance Segmentation

This work proposes 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.

Matting Enhanced Mask R-CNN

A novel and effective method for high-quality instance segmentation using a novel matting enhanced mask head to generate trimap-based mat-ting features as auxiliary priors and a Uncertainty-Aware Binary Cross-Entropy Loss to assign larger weights to pixels with higher uncertainty.

SOLO: Segmenting Objects by Locations

A new, embarrassingly simple approach to instance segmentation in images by introducing the notion of "instance categories", which assigns categories to each pixel within an instance according to the instance's location and size thus nicely converting instance mask segmentation into a classification-solvable problem.

PointRend: Image Segmentation As Rendering

The PointRend (Point-based Rendering) neural network module is presented: a module that performs point-based segmentation predictions at adaptively selected locations based on an iterative subdivision algorithm that enables output resolutions that are otherwise impractical in terms of memory or computation compared to existing approaches.

End-to-End Video Instance Segmentation with Transformers

A new video instance segmentation framework built upon Transformers, termed VisTR, which views the VIS task as a direct end-to-end parallel sequence decoding/prediction problem, and achieves the highest speed among all existing VIS models and the best result among methods using single model on the YouTube-VIS dataset.