Faculty Profiles - Deguchi Daisuke

Presentations -

Division display 1 - 20 of about 713 ／ All the affair displays >>

講義スライドのコンテンツ解析に基づく講義動画の自動生成に関する予備検討

水上皓太, 出口大輔, 久徳遙矢, 峰松翼, 村瀬洋

教育学習支援情報システム研究会(CLE) 2025.3.21

Multi-group Vision Semantic Centroid for Semantic Segmentation International conference

Jialei Chen, Daisuke Deguchi, Chenkai Zhang, Zhenzhen Quan, Seigo Ito, Hiroshi Murase

Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV2025) 2025.2.18

Event date： 2025

Language：English Presentation type：Oral presentation (general)

Venue：Sogang University, Seoul, Korea

Multi-media processing has achieved great success based on
semantic segmentation. Semantic segmentation can be viewed as pixel-
clustering based on semantic prototypes. However, existing methods fo-
cus more on consistent semantics while ignoring the consistency in vi-
sion, making this task challenging. Motivated by the success of discrete
visual representation learning, we propose Multi-group Visual Semantic
Centroid (MVSC) to better cluster the pixels while maintaining consis-
tent semantics of the dense features for any image encoder. Specifically,
we randomly initialize multiple groups of prototypes as multi-groups in
visual space. The visual features are also randomly split into the same
groups and forced to be aligned with the corresponding prototypes. Then
these visual prototypes are projected into the semantic space and super-
vised by the same classifier as the dense features. Compared with existing
methods, MVSC further considers the visual space and thus facilitates
the task. Experimental results on COCO-Stu! show great improvements
compared with previous methods.

A Cross-Modal Knowledge Distillation Approach for RGB-to-Infrared Video Action Recognition International conference

Zhenzhen Quan, Daisuke Deguchi, Jialei Chen, Chenkai Zhang, Yujun Li, Seigo Ito, Hiroshi Murase

Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV2025) 2025.2.18

Object Detection using In-Vehicle Camera Images with Object Attributes Constraints

星谷那月, Daisuke Deguchi, Jialei Chen, Hiroshi Murase, 伊藤誠悟

IEICE Technical Report (MVE) 2025.3.5

複数歩行者間のインタラクション理解に向けた気づき認識データセットの構築

辰己弘征, 出口大輔, 村瀬洋, 伊藤誠悟

メディアエクスペリエンス・バーチャル環境基礎研究会(MVE) 2025.3.5

学習活動ログと講義スライドの併用による早期At-risk 検出

尾﨑優也, 出口大輔, 久徳遙矢, 村瀬洋

教育学習支援情報システム研究会(CLE) 2025.3.20

A preliminary study on colorized NeRF with grayscale images.

水上皓太, Daisuke Deguchi, Hiroshi Murase

2024.3.17

A Preliminary Study on Pre-training for Improved Explainability in End-to-End Driving Models

Chenkai Zhang, Daisuke Deguchi, Jialei Chen, Hiroshi Murase

Pattern Recognition and Media Understanding (PRMU) 2024.1.26

Event date： 2024

Language：English Presentation type：Oral presentation (general)

Venue：Keio University

In the realm of autonomous driving, end-to-end models (E2EDMs) have gained prominence due to their high predictive accuracy. Such accuracy is attributed to the utilization of a backbone pre-trained on large datasets, subsequently fine-tuned on autonomous driving datasets. However, the inherent “black box” nature of these E2EDMs poses significant challenges in terms of explainability. Current methodologies predominantly focus on generating visual explanations for the E2EDMs’ decision-making process. Numerous approaches aim to enhance the explainability of these E2EDMs by fine-tuning with complicated architectures, supplemented by additional information, e.g., object position to develop more explainable E2EDMs. In this study, we diverge from the conventional approaches where significant effort is placed during the fine-tuning phase of E2EDMs. Our method focuses on training backbones before the fine-tuning phase. This preemptive strategy enables us to fine-tune more explainable E2EDMs without the need for additional information or complex training techniques.

Pedestrian's Gaze Object Detection in Traffic Scene International conference

Hiroto Murakami, Jialei Chen, Daisuke Deguchi, Takatsugu Hirayama, Yasutomo Kawanishi, Hiroshi Murase

19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP) 2024 2024.2.27

Event date： 2024

Language：English Presentation type：Poster presentation

Venue：Rome, Italy

In this paper, we present a new task of detecting an object that a target pedestrian is gazing at in a traffic scene called PEdestrian's Gaze Object (PEGO). We argue that the detection of gaze object can provide important information for pedestrian's behavior prediction and can contribute to the realization of automated vehicles. For this task, we construct a dataset of in-vehicle camera images with annotations of the objects that pedestrians are gazing at. Also, we propose a Transformer-based method called PEGO Transformer to solve the PEGO detection task. The PEGO Transformer directly performs gaze object detection with the utilization of whole-body features without a high-resolution head image and a gaze heatmap which the traditional methods rely on. Experimental results showed that the proposed method could estimate pedestrian's gaze object accurately even if various objects exist in the scene.

Comprehensive Evaluation of End-to-End Driving Model Explanations for Autonomous Vehicles International conference

Chenkai Zhang, Daisuke Deguchi, Jialei Chen, Hiroshi Murase

19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP) 2024 2024.2.27

Event date： 2024

Language：English Presentation type：Oral presentation (general)

Venue：Precise House Mantegna Roma

Deep learning technology has rapidly advanced, leading to the development of End-to-End driving models (E2EDMs) for autonomous vehicles with high prediction accuracy. To comprehend the prediction results of these E2EDMs, one of the most representative explanation methods is attribution-based. There are two kinds of attribution-based explanation methods: pixel-level and object-level. Usually, the heatmaps illustrate the importance of pixels and objects in the prediction results, serving as explanations for E2EDMs. Since there are many attribution-based explanation methods, evaluation methods are proposed to determine which one is better at improving the explainability of E2EDMs.
Fidelity measures the explanation's faithfulness to the model's prediction method, which is a bottommost property. However, no evaluation method could measure the fidelity difference between object-level and pixel-level explanations, making the current evaluation incomplete. In addition, without considering fidelity, previous evaluation methods may advertise manipulative explanations that solely seek human satisfaction (persuasibility). Therefore, we propose an evaluation method that further considers fidelity, our method enables a comprehensive evaluation that proves the object-level explanations genuinely outperform pixel-level explanations in fidelity and persuasibility, thus could better improve the explainability of the E2EDMs.

交通シーン画像からの歩行者の注視対象物推定

村上大斗, 陳嘉雷, 出口大輔, 平山高嗣, 川西康友, 村瀬洋

動的画像処理実利用化ワークショップ(DIA) 2024.3.4

Event date： 2024

Language：Japanese Presentation type：Oral presentation (general)

Venue：別府国際コンベンションセンタービーコンプラザ

注視対象物推定は，自動運転車両の実現において不可欠な歩行者の行動予測のための重要な要素である．本発表では，交通シーンで歩行者が注視している物体（PEdestrians' Gaze Object: PEGO）を推定するPEGO Transformer V2を提案する．従来手法の多くは，高解像度の頭部画像と視線ヒートマップを用いて歩行者の注視対象物を推定している．しかしながら，遠方にいる歩行者に対して，高解像度の頭部画像を得られないため推定精度が低いという問題があった．そこでPEGO Transformer V2は，歩行者全身の特徴に加え，歩行者位置情報も活用することで性能改善を図る．実験の結果，提案手法であるPEGO Transformer V2は，シーン中の歩行者ごとにそれぞれ正しい注視対象物を推定できることを確認した．

A preliminary study on front image generation from back by diffusion model

松雄なずな, Daisuke Deguchi, Hiroshi Murase

2024.3.17

A Study on Fixed Orthogonal Prototype Classifier for Semantic Segmentation

Jialei Chen, Daisuke Deguchi, Chenkai Zhang, Xu Zheng, Hiroshi Murase

Pattern Recognition and Media Understanding (PRMU) 2024.1.26

Distill from CLIP by Token Alignment for Inductive Zero-shot Semantic Segmentation

Jialei Chen, Daisuke Deguchi, Chenkai Zhang, Zhenzhen Quan, Hiroshi Murase

2024.8.9

Enhancing Explainability of End-to-End Autonomous Driving Models through additional fine-tuning

Chenkai Zhang, Daisuke Deguchi, Jialei Chen, Hiroshi Murase

27th Meeting on Image Recognition and Understanding (MIRU2024) 2024.8.9

Event date： 2024

Language：English Presentation type：Poster presentation

Venue：熊本城ホール（熊本県熊本市中央区桜町3番40号）

In autonomous driving, end-to-end driving models (E2EDMs) have become increasingly popular due to their exceptional predictive performance. This performance stems from leveraging a backbone that is pre-trained on large datasets and then fine-tuned for driving tasks. However, the opaque nature of these E2EDMs presents a challenge to explainability, thus visual explanations are created to shed light on the decision-making process of E2EDMs. Currently, these E2EDMs have been made more explainable by fine-tuning the driving tasks along with a side task designed for explainability. This side task for explainability requires complex structures with auxiliary data, such as the location of objects.
In this paper, we argue that a more effective approach to enhancing the explainability of E2EDMs is to design a fine-tuning specifically focused on explainability. We propose CROp-based COntrastive DIscriminative LEarning (CROCODILE) to enhance a backbone’s capability to accurately identify object features during this additional fine-tuning for explainability. By adopting CROCODILE, we can develop more explainable E2EDMs without relying on auxiliary data or complex structures in the fine-tuning for driving tasks. Our experimental results confirm that our approach enhances the explainability of E2EDMs.

A Decoupled Knowledge Distillation Approach for RGB-to-Infrared Video Action Recognition

Zhenzhen Quan, Jialei Chen, Chenkai Zhang, Daisuke Deguchi, Hiroshi Murase

27th Meeting on Image Recognition and Understanding (MIRU2024) 2024.8.9

学習活動ログを用いたAt-risk 学生の早期検出：学習活動予測ネットワークの活用

尾﨑優也, 出口大輔, 久徳遙矢, 村瀬洋

第27回画像の認識・理解シンポジウム(MIRU2024) 2024.8.9

Front Image Generation from Back by Diffusion Model

Nazuna Matsuo, Daisuke Deguchi, Hiroshi Murase

27th Meeting on Image Recognition and Understanding (MIRU2024) 2024.8.9

Centroid Module for Shaping Feature Space in Semantic Segmentation International conference

Jialei Chen, Daisuke Deguchi, Chenkai Zhang, Hiroshi Murase

3rd Asia Conference on Cloud Computing, Computer Vision and Image Processing (3CVIP 2024) 2024.3.23

A construction of colorized NeRF with grayscale images

Kota Mizukami, Daisuke Deguchi, Hiroshi Murase

27th Meeting on Image Recognition and Understanding (MIRU2024) 2024.8.6

　 PREV - NEXT 　

<Deguchi Daisuke>

Presentations -