Amodal video instance segmentation

We explore approaches to improve over existing amodal prediction models for the task of semantic amodal instance level video object segmentation, i.e., the task to delineate objects and their occluded parts in video data. We propose Amodal-Net with three improvements: First, we leverage temporal information. Specifically, we employ 3D convolutions and a flow alignment module which permits to aggregate the objects’ features across frames. Second, we develop a cascaded box-head with soft-non-maximum-suppression to address the challenge that amodal segmentations overlap significantly. Third, we address the challenge that occlusions require observation information to be propagated over larger distances by developing an attention-based mask-head. Then we also study reprojection, another way of using temporal information which also uses 3D information. We evaluate our approach on amodal segmentation for video data, SAILVOS.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2022

Author dc:creator

• Sun, Mingxi

Contributors dc:contributor

• Schwing, Alexander

• machine learning
• computer vision
• instance segmentation
• amodal segmentation

Statement dc:rights

• Copyright 2021 Mingxi Sun

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/113229

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/113229