Presentation 2023-01-27
Importance ranking estimation of objects in a city scene
Yasuhiro Nitta, Mariko Isogawa, Ryo Yonetani, Maki Sugimoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recent studies have been conducted on visual assistance such as object recognition in urban scenes, to assist visually impaired people. These studies have achieved high accuracy in recognizing objects such as obstacles. On the other hand, there is room for further study on the order prediction in which information should be provided.In this study, we trained the importance ranking estimation network for the detected obstacles to clarify how to convey surrounding information to a visually impaired person. We input the optical flow image from the previous frame and the region information of the detected objects as input. It allowed us to learn the network considering the distance from the user and the direction of movement. Experimental results showed that the model trained by this method had higher ranking estimation accuracy than the three baseline methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Support for the Visually Impaired Person / Differentiable Sorting / Object Detection
Paper # MVE2022-42
Date of Issue 2023-01-19 (MVE)

Conference Information
Committee MVE / IPSJ-CVIM / VRSJ-SIG-MR
Conference Date 2023/1/26(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Nara Institute of Science and Technology
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Kiyoshi Kiyokawa(NAIST)
Vice Chair Sumaru Niida(KDDI Research)
Secretary Sumaru Niida(NAIST) / (DNP) / (Univ. of ToKyo)
Assistant Hidehiko Shishido(Univ. of Tsukuba) / Atsushi Nakazawa(Kyoto Univ.) / Naoya Tojo(KDDI Research) / Naoki Hagiyama(NTT)

Paper Information
Registration To Technical Committee on Media Experience and Virtual Environment / Special Interest Group on Computer Vision and Image Media / SIG-MR
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Importance ranking estimation of objects in a city scene
Sub Title (in English)
Keyword(1) Support for the Visually Impaired Person
Keyword(2) Differentiable Sorting
Keyword(3) Object Detection
1st Author's Name Yasuhiro Nitta
1st Author's Affiliation Keio University(Keio Univ.)
2nd Author's Name Mariko Isogawa
2nd Author's Affiliation Keio University/JET,Presto(Keio Univ./Presto)
3rd Author's Name Ryo Yonetani
3rd Author's Affiliation Keio University(Keio Univ.)
4th Author's Name Maki Sugimoto
4th Author's Affiliation Keio University(Keio Univ.)
Date 2023-01-27
Paper # MVE2022-42
Volume (vol) vol.122
Number (no) MVE-367
Page pp.pp.43-48(MVE),
#Pages 6
Date of Issue 2023-01-19 (MVE)