Abstract
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.
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 language | English |
---|---|
Title of host publication | Reflective Field for Pixel-Level Tasks |
Publisher | IEEE, Institute of Electrical and Electronics Engineers |
Pages | 529-534 |
Number of pages | 6 |
ISBN (Electronic) | 9781538637883 |
ISBN (Print) | 9781538637883 |
DOIs | |
Publication status | Published - 26 Nov 2018 |
Event | 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China Duration: 20 Aug 2018 → 24 Aug 2018 |
Publication series
Name | Proceedings - International Conference on Pattern Recognition |
---|---|
Volume | 2018-August |
ISSN (Print) | 1051-4651 |
Conference
Conference | 2018 24th International Conference on Pattern Recognition (ICPR) |
---|---|
Country/Territory | China |
City | Beijing |
Period | 20/08/18 → 24/08/18 |