Plant Leaf Segmentation Using Knowledge Distillation

Joo Yeon Jung; Sang Ho Lee; Jong Ok Kim

doi:10.1109/ICCE-Asia57006.2022.9954844

Plant Leaf Segmentation Using Knowledge Distillation

Joo Yeon Jung
, Sang Ho Lee
, Jong Ok Kim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Citations (Scopus)

Abstract

This paper proposes a method to segment plant leaves using knowledge distillation. Unlike the existing knowledge distillation method aimed at lightening the model, the architectures of the teacher and student networks are kept identical. Plants have many leaves, and each leaf is very small. To segment each plant leaf well, clustering is used through spatial embedding. The teacher and student networks perform segmentation based on spatial embedding. The teacher network is trained with a large dataset and then distills its segmentation knowledge into the student network. Two types of knowledge are distilled from the teacher network: feature distillation and attention distillation. The results of the experiment demonstrate that better instance segmentation can be achieved when using knowledge distillation.

Original language	English
Title of host publication	2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665464345
DOIs	https://doi.org/10.1109/ICCE-Asia57006.2022.9954844
Publication status	Published - 2022
Event	2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022 - Yeosu, Korea, Republic of Duration: 2022 Oct 26 → 2022 Oct 28

Publication series

Name	2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022

Conference

Conference	2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022
Country/Territory	Korea, Republic of
City	Yeosu
Period	22/10/26 → 22/10/28

Bibliographical note

Funding Information:
This work is supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2020R1A4A4079705).

Publisher Copyright:
© 2022 IEEE.

Keywords

knowledge distillation
leaf instance segmentation
spatial embedding

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering
Media Technology
Instrumentation

Access to Document

10.1109/ICCE-Asia57006.2022.9954844

Cite this

Jung, J. Y., Lee, S. H., & Kim, J. O. (2022). Plant Leaf Segmentation Using Knowledge Distillation. In 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022 (2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCE-Asia57006.2022.9954844

Plant Leaf Segmentation Using Knowledge Distillation. / Jung, Joo Yeon; Lee, Sang Ho; Kim, Jong Ok.
2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Jung, JY, Lee, SH & Kim, JO 2022, Plant Leaf Segmentation Using Knowledge Distillation. in 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022. 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022, Institute of Electrical and Electronics Engineers Inc., 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022, Yeosu, Korea, Republic of, 22/10/26. https://doi.org/10.1109/ICCE-Asia57006.2022.9954844

@inproceedings{aef6cabbc10045459cae90d449e01bdf,

title = "Plant Leaf Segmentation Using Knowledge Distillation",

abstract = "This paper proposes a method to segment plant leaves using knowledge distillation. Unlike the existing knowledge distillation method aimed at lightening the model, the architectures of the teacher and student networks are kept identical. Plants have many leaves, and each leaf is very small. To segment each plant leaf well, clustering is used through spatial embedding. The teacher and student networks perform segmentation based on spatial embedding. The teacher network is trained with a large dataset and then distills its segmentation knowledge into the student network. Two types of knowledge are distilled from the teacher network: feature distillation and attention distillation. The results of the experiment demonstrate that better instance segmentation can be achieved when using knowledge distillation.",

keywords = "knowledge distillation, leaf instance segmentation, spatial embedding",

author = "Jung, \{Joo Yeon\} and Lee, \{Sang Ho\} and Kim, \{Jong Ok\}",

note = "Funding Information: This work is supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2020R1A4A4079705). Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022 ; Conference date: 26-10-2022 Through 28-10-2022",

year = "2022",

doi = "10.1109/ICCE-Asia57006.2022.9954844",

language = "English",

series = "2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022",

}

TY - GEN

T1 - Plant Leaf Segmentation Using Knowledge Distillation

AU - Jung, Joo Yeon

AU - Lee, Sang Ho

AU - Kim, Jong Ok

N1 - Funding Information: This work is supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2020R1A4A4079705). Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - This paper proposes a method to segment plant leaves using knowledge distillation. Unlike the existing knowledge distillation method aimed at lightening the model, the architectures of the teacher and student networks are kept identical. Plants have many leaves, and each leaf is very small. To segment each plant leaf well, clustering is used through spatial embedding. The teacher and student networks perform segmentation based on spatial embedding. The teacher network is trained with a large dataset and then distills its segmentation knowledge into the student network. Two types of knowledge are distilled from the teacher network: feature distillation and attention distillation. The results of the experiment demonstrate that better instance segmentation can be achieved when using knowledge distillation.

AB - This paper proposes a method to segment plant leaves using knowledge distillation. Unlike the existing knowledge distillation method aimed at lightening the model, the architectures of the teacher and student networks are kept identical. Plants have many leaves, and each leaf is very small. To segment each plant leaf well, clustering is used through spatial embedding. The teacher and student networks perform segmentation based on spatial embedding. The teacher network is trained with a large dataset and then distills its segmentation knowledge into the student network. Two types of knowledge are distilled from the teacher network: feature distillation and attention distillation. The results of the experiment demonstrate that better instance segmentation can be achieved when using knowledge distillation.

KW - knowledge distillation

KW - leaf instance segmentation

KW - spatial embedding

UR - https://www.scopus.com/pages/publications/85143847369

U2 - 10.1109/ICCE-Asia57006.2022.9954844

DO - 10.1109/ICCE-Asia57006.2022.9954844

M3 - Conference contribution

AN - SCOPUS:85143847369

T3 - 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022

BT - 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022

Y2 - 26 October 2022 through 28 October 2022

ER -

Plant Leaf Segmentation Using Knowledge Distillation

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this