ken-system: Advance Program

IEICE Technical Committee Submission System
Advance Program

Online Proceedings
[Sign in]
Tech. Rep. Archives

===============================================
Technical Committee on Pattern Recognition and Media Understanding (PRMU)
Chair: Seiichi Uchida (Kyushu Univ.) Vice Chair: Takuya Funatomi (NAIST), Mitsuru Anpai (Denso IT Lab.)
Secretary: Kouta Yamaguchi (CyberAgent), Yusuke Matsui (Univ. of Tokyo)
Assistant: Nakamasa Inoue (Tokyo Inst. of Tech.), Yasutomo Kawanishi (Riken)

DATE:
Thu, Dec 15, 2022 10:00 - 16:40
Fri, Dec 16, 2022 10:00 - 17:45

PLACE:

TOPICS:

----------------------------------------
Thu, Dec 15 AM (10:00 - 11:30)
----------------------------------------

(1) 10:00 - 10:15
Development of a virtual aerial display that can interact with 3DCG objects
Hideaki Saito, Hiroki Imamura (Soka Univ)

(2) 10:15 - 10:30
Multi object tracking using multiple frames features.
Takanori Asanomi, Kazuya Nishimura, Ryoma Bise (Kyushu Univ.)

(3) 10:30 - 10:45
Single Image Raindrop Removal Using a Non-local Operator and Feature Maps in the Frequency Domain
Shinya Ezumi, Masaaki Ikehara (Keio Univ.)

(4) 10:45 - 11:00
DN4C
-- An Interactive Image Segmentation System Combining Deep Neural Network and Nearest Neighbor Classifier --
Toshikazu Wada, Koji Kamma (Wakayama University)

(5) 11:00 - 11:30

----- Break ( 80 min. ) -----

----------------------------------------
Thu, Dec 15 PM (12:50 - 13:50)
----------------------------------------

(6) 12:50 - 13:50

----- Break ( 10 min. ) -----

----------------------------------------
Thu, Dec 15 PM (14:00 - 15:20)
----------------------------------------

(7) 14:00 - 14:10
(See Japanese page.)

(8) 14:10 - 14:25
*
Yuki Taya, Rin Takano, Hiroyuki Oyama (NEC)

(9) 14:25 - 14:40
A DNN compression method based on output error of activation functions
Koji Kamma, Toshikazu Wada (Wakayama Univ.)

(10) 14:40 - 14:50
[Short Paper]
A Module Selection-based Approach for Efficient Skeleton Human Action Recognition
Shurong Chai, Rahul Kumar Jain, Jiaqing Liu (Ritsumeikan Univ.), Tomoko Tateyama (Fujita Health Univ), Yinhao Li, Yen-Wei Chen (Ritsumeikan Univ.)

(11) 14:50 - 15:20

----- Break ( 10 min. ) -----

----------------------------------------
Thu, Dec 15 PM (15:30 - 16:40)
----------------------------------------

(12) 15:30 - 15:45
Training Method for Image-based Instance Segmentation by Video-based Object-Centric Representation Learning
Tomokazu Kaneko, Ryosuke Sakai, Soma Shiraishi (NEC)

(13) 15:45 - 16:00
Implementation of Intuitive 3-Dmensional Manipulation for 3DCG Objects Using Monocular Camera
Hibiki Ito, Hiroki Imamura (Souka Univ.)

(14) 16:00 - 16:10
[Short Paper]
An Intra- and Inter-Modality Transformer-based Fusion Model Using Multiemotional Audiovisual Features for Depression Prediction
Shiyu Teng, Shurong Chai, Jiaqing Liu (Ritsumeikan Univ.), Tateyama Tomoko (Fujita Health Univ.), Xinyin Huang (Soochow Univ.), Yen-wei Chen (Ritsumeikan Univ.)

(15) 16:10 - 16:40

----------------------------------------
Fri, Dec 16 AM (10:00 - 11:40)
----------------------------------------

(16) 10:00 - 10:15
Image Transformation Simulating the Height Change of Wide-angle Ceiling Camera for People Detection
Hayato Kitao, Toshikazu Wada (Wakayama Univ.)

(17) 10:15 - 10:30
Pose-aware Disentangled Multiscale Transformer for Pose Guided Person Image Generation
Kei Shibasaki, Masaaki Ikehara (Keio Univ.)

(18) 10:30 - 10:45
Digitization of education workbooks and collection of handwritten answers toward their automatic scoring
Tomo Asakura (Wacom), Nguyen Tuan Hung, Nguyen Tuan Cuong (Tokyo Univ. of Agri. and Tech.), Hiroshi Miyazawa, Yoichi Tsuchida, Takahiro Yamamoto, Masamitsu Ito, Toshihiko Horie (Wacom), Ikuko Shimizu, Masaki Nakagawa (Tokyo Univ. of Agri. and Tech.)

(19) 10:45 - 11:00
Development of a virtual aerial display that can interact with 3DCG objects
Koji Nammatsu, Hiroki Imamura (Soka Univ)

(20) 11:00 - 11:10
[Short Paper]
Contrastive Learning using Adversarial Metric Learning Modules for Robust Image Recognition
Yoshifumi Sugiyama (Tokyo Tech), Hirokatsu Kataoka (AIST), Rio Yokota, Nakamasa Inoue (Tokyo Tech)

(21) 11:10 - 11:40

----- Break ( 80 min. ) -----

----------------------------------------
Fri, Dec 16 PM (13:00 - 14:00)
----------------------------------------

(22) 13:00 - 14:00

----- Break ( 10 min. ) -----

----------------------------------------
Fri, Dec 16 PM (14:10 - 15:50)
----------------------------------------

(23) 14:10 - 14:25
Sampling Strategies in Data Pruning
Ryota Higashi, Toshikazu Wada (Wakayama Univ.)

(24) 14:25 - 14:40
Investigation for the Importance of Brightness Adjustment in Image Recognition for Water Hyacinth Using Semantic Segmentation
Soma Miyamoto, Hiroki Imamura (Soka Univ.)

(25) 14:40 - 14:55
Data Augmentation
Shumpei Takezaki (Kyushu Univ.), Kiyohito Tanaka (Kyoto Second Red Cross Hospital), Seiichi Uchida, Takeaki Kadota (Kyushu Univ.)

(26) 14:55 - 15:10
On a difficulty of TextVQA
Koki Nakamura, Seiichi Uchida (Kyushu Univ.)

(27) 15:10 - 15:20
[Short Paper]
Cosine Similarity Based Attention on a Hypersphere for Vision Transformers
Jungdae Lee, Rei Kawakami, Nakamasa Inoue (Tokyo Tech)

(28) 15:20 - 15:50

----- Break ( 10 min. ) -----

----------------------------------------
Fri, Dec 16 PM (16:00 - 17:45)
----------------------------------------

(29) 16:00 - 16:15
Extraction of Feature Flow for Deforming Non-rigid Objects and Its Analysis Method
Yuki Fukuda, Munetoshi Iwakiri, Takumi Fujiwara (NDA)

(30) 16:15 - 16:30
On partitioning of the space of one-dimensional normal distributions based on e(m)-divergence
Jun Fujiki (Fukuoka Univ.), Shotaro Akaho (AIST)

(31) 16:30 - 16:45
Cell Detection from Partially Annotated Data
Kazuma Fujii (Kyushu Univ.), Daiki suehiro (Kyushu Univ./AIP), Ryouma Bise (Kyushu Univ.)

(32) 16:45 - 17:00
Training Kindai OCR with parallel textline images and self-attention feature distance-based loss
Le Duc Anh, Kitamoto Asanobu (Center for Open Data in the Humanities)

(33) 17:00 - 17:15
learning label proportions
Shinnosuke Matsuo, Ryoma Bise, Seiichi Uchida, Daiki Suehiro (Kyushu Univ.)

(34) 17:15 - 17:45

# Information for speakers
General Talk will have 15 minutes for presentation.
General Talk (ショートペーパー) will have 10 minutes for presentation.

=== Technical Committee on Pattern Recognition and Media Understanding (PRMU) ===
# FUTURE SCHEDULE:

Thu, Mar 2, 2023 - Fri, Mar 3, 2023: Future University Hakodate [Thu, Jan 5]

# SECRETARY:
Contact address: pr