Instance and Panoptic Segmentation Using Conditional Convolutions

02/05/2021
by   Zhi Tian, et al.
0

We propose a simple yet effective framework for instance and panoptic segmentation, termed CondInst (conditional convolutions for instance and panoptic segmentation). In the literature, top-performing instance segmentation methods typically follow the paradigm of Mask R-CNN and rely on ROI operations (typically ROIAlign) to attend to each instance. In contrast, we propose to attend to the instances with dynamic conditional convolutions. Instead of using instance-wise ROIs as inputs to the instance mask head of fixed weights, we design dynamic instance-aware mask heads, conditioned on the instances to be predicted. CondInst enjoys three advantages: 1.) Instance and panoptic segmentation are unified into a fully convolutional network, eliminating the need for ROI cropping and feature alignment. 2.) The elimination of the ROI cropping also significantly improves the output instance mask resolution. 3.) Due to the much improved capacity of dynamically-generated conditional convolutions, the mask head can be very compact (e.g., 3 conv. layers, each having only 8 channels), leading to significantly faster inference time per instance and making the overall inference time almost constant, irrelevant to the number of instances. We demonstrate a simpler method that can achieve improved accuracy and inference speed on both instance and panoptic segmentation tasks. On the COCO dataset, we outperform a few state-of-the-art methods. We hope that CondInst can be a strong baseline for instance and panoptic segmentation. Code is available at: https://git.io/AdelaiDet

READ FULL TEXT

page 3

page 4

page 10

page 11

research
03/12/2020

Conditional Convolutions for Instance Segmentation

We propose a simple yet effective instance segmentation framework, terme...
research
05/29/2021

FCPose: Fully Convolutional Multi-Person Pose Estimation with Dynamic Instance-Aware Convolutions

We propose a fully convolutional multi-person pose estimation framework ...
research
07/15/2022

3D Instances as 1D Kernels

We introduce a 3D instance representation, termed instance kernels, wher...
research
03/23/2020

SOLOv2: Dynamic, Faster and Stronger

In this work, we aim at building a simple, direct, and fast instance seg...
research
06/28/2021

K-Net: Towards Unified Image Segmentation

Semantic, instance, and panoptic segmentations have been addressed using...
research
03/18/2021

SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentation

Video instance segmentation (VIS) is a new and critical task in computer...
research
12/09/2021

Implicit Feature Refinement for Instance Segmentation

We propose a novel implicit feature refinement module for high-quality i...

Please sign up or login with your details

Forgot password? Click here to reset