Reflective Field for Pixel-Level Tasks

Liang Zhang, X Kong, Peiyi Shen, Guangming Zhu, Syed Shah, Mohammed Bennamoun, J. Song

Research output: Chapter in Book/Conference paperConference paperpeer-review


PixelNet has achieved great success in dense prediction
problems with a pure pixel-level architecture, but there is
still much room for improvement. In this paper, we start from
PixelNet and discuss the pixel-level architecture called hypercolumn
and its limitations in building feature representation with
rich semantic information. To achieve this goal, we propose a
concept in the context of neural networks called reflective field,
representing the area reflected by the origin input. Furthermore,
the proposed reflective field is used to solve the limitations of
the hypercolumn architecture. Specifically, we give the method of
calculating the size of the reflective field and analyze the effective
reflective field in the calculated area. Then, we use the reflective
field to build a new hypercolumn architecture, which has a more
rational construction. The results on PASCAL VOC segmentation
dataset with our new architecture are improved.
Original languageEnglish
Title of host publicationReflective Field for Pixel-Level Tasks
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)9781538637883
ISBN (Print)9781538637883
Publication statusPublished - 26 Nov 2018
Event2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China
Duration: 20 Aug 201824 Aug 2018

Publication series

NameProceedings - International Conference on Pattern Recognition
ISSN (Print)1051-4651


Conference2018 24th International Conference on Pattern Recognition (ICPR)


Dive into the research topics of 'Reflective Field for Pixel-Level Tasks'. Together they form a unique fingerprint.

Cite this