Robust Instance Segmentation through Reasoning about Multi-Object Occlusion

12/03/2020
by   Xiaoding Yuan, et al.
9

Analyzing complex scenes with Deep Neural Networks is a challenging task, particularly when images contain multiple objects that partially occlude each other. Existing approaches to image analysis mostly process objects independently and do not take into account the relative occlusion of nearby objects. In this paper, we propose a deep network for multi-object instance segmentation that is robust to occlusion and can be trained from bounding box supervision only. Our work builds on Compositional Networks, which learn a generative model of neural feature activations to locate occluders and to classify objects based on their non-occluded parts. We extend their generative model to include multiple objects and introduce a framework for the efficient inference in challenging occlusion scenarios. In particular, we obtain feed-forward predictions of the object classes and their instance and occluder segmentations. We introduce an Occlusion Reasoning Module (ORM) that locates erroneous segmentations and estimates the occlusion ordering to correct them. The improved segmentation masks are, in turn, integrated into the network in a top-down manner to improve the image classification. Our experiments on the KITTI INStance dataset (KINS) and a synthetic occlusion dataset demonstrate the effectiveness and robustness of our model at multi-object instance segmentation under occlusion.

READ FULL TEXT

page 1

page 3

page 4

page 5

page 6

page 7

research
10/25/2020

Weakly-Supervised Amodal Instance Segmentation with Compositional Priors

Amodal segmentation in biological vision refers to the perception of the...
research
11/25/2020

CellSegmenter: unsupervised representation learning and instance segmentation of modular images

We introduce CellSegmenter, a structured deep generative model and an am...
research
03/12/2023

Amodal Intra-class Instance Segmentation: New Dataset and Benchmark

Images of realistic scenes often contain intra-class objects that are he...
research
03/05/2022

Towards Robust Part-aware Instance Segmentation for Industrial Bin Picking

Industrial bin picking is a challenging task that requires accurate and ...
research
12/08/2016

Deep Supervision with Shape Concepts for Occlusion-Aware 3D Object Parsing

Monocular 3D object parsing is highly desirable in various scenarios inc...
research
02/25/2021

Blocks World Revisited: The Effect of Self-Occlusion on Classification by Convolutional Neural Networks

Despite the recent successes in computer vision, there remain new avenue...
research
04/06/2020

Self-Supervised Scene De-occlusion

Natural scene understanding is a challenging task, particularly when enc...

Please sign up or login with your details

Forgot password? Click here to reset