研究者詳細 - 出口　大輔

論文 - 出口　大輔

分割表示 253 件中 21 - 40 件目／全件表示 >>

More Persuasive Explanation Method for End-to-End Driving Models 査読有り

Chenkai Zhang, Daisuke Deguchi, Yuki Okafuji, Hiroshi Murase

IEEE Access 11 巻頁： 4270 - 4282 2023年1月

記述言語：英語掲載種別：研究論文（学術雑誌）

With the rapid development of autonomous driving technology, a variety of high-performance end-to-end driving models (E2EDMs) are being proposed. In order to understand the computational methods of E2EDMs, pixel-level explanations methods are used to obtain the explanations of the E2EDMs. However, little attention has been paid to the excellence of the explanations of E2EDMs. Therefore, in order to build trustworthy E2EDMs, we focus on improving the persuasibility of the explanations of E2EDMs. We propose an object-level explanation method (main approach) for E2EDMs, which masks the objects in the image and then treats the change in the prediction result as the importance of the objects, then we explain the E2EDM by the importance of each object. To further validate the effectiveness of object-level explanations, we propose another approach (validation approach), which trains E2EDMs with object information as input and generates the importance of objects using general explanation methods. Both approaches generate object-level explanations, in order to compare these object-level explanations with traditional pixel-level explanations, we propose experimental methods to measure the persuasibility of explanations of E2EDMs through a subjective and objective method. The subjective method evaluates persuasibility based on the extent to which participants think the importance of features indicated by the explanations is correct. The objective method evaluates the persuasibility based on the human annotation similarity between provided with only the important part of images and provided with the complete images. The experimental results show that the object-level explanations are more persuasive than the traditional pixel-level explanations.

DOI： 10.1109/ACCESS.2023.3235739

Analysis of Gaze while Viewing Videos at Various Playback Speeds 査読有り

Hideaki Takahira, Yasutomo Kawanishi, Takatsugu Hirayama, Hiroshi Murase, Daisuke Deguchi

Proceedings of the International Display Workshops 29 巻頁： 702 - 705 2022年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We analyze gaze while viewing videos at various playback speeds.
The aims of this paper are to evaluate the comprehension and viewability of videos.
In our experiment, we used 0.75x to 3.0x speed, and measured gaze.
We report differences in gaze behavior among different playback speeds in viewing same video.

Detection of Birds in a 3D Environment Referring to Audio-Visual Information 査読有り

Yasutomo Kawanishi, Ichiro Ide, Baidong Chu, Chihaya Matsuhira, Marc A. Kastner, Takahiro Komamizu, Daisuke Deguchi

Proceedings of the 18th IEEE International Conference on Advanced Video and Signal-Based Surveillance 頁： 1 - 7 2022年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We propose a method to detect birds in a 3D environment referring to both audio information observed from a microphone array and visual information observed from a panorama camera. In general, in panorama images, birds appear relatively too small to be detected accurately even with the state-of-the-art deep learning models. Thus, the proposed method takes a two step approach where the birds are first roughly located referring to audio information by Sound Source Localization (SSL), and then image detection is applied within its vicinity. Through evaluation on a dataset annotated with bounding boxes surrounding the birds, we show that the proposed method improves detection performance of birds that appear in relatively small sizes in the image, in both accuracy and processing speed.

DOI： 10.1109/AVSS56176.2022.9959510

Detection of distant eye-contact using spatio-temporal pedestrian skeletons 査読有り

Ryusei Hata, Daisuke Deguchi, Takatsugu Hirayama, Yasutomo Kawanishi, Hiroshi Murase

2022年9月

サラウンド映像視聴時の注視点分布の分析査読有り

高比良英朗, 篠原未歩, 野坂祐介, 横内雅也, 山田光穗, 平山高嗣, 川西康友, 出口大輔, 井手一郎, 村瀬洋

ヒューマンインタフェース学会論文誌 24 巻 ( 3 ) 頁： 195 - 204 2022年8月

記述言語：日本語掲載種別：研究論文（学術雑誌）

We have been conducting research focusing on the human gaze as a principal biological reaction while viewing videos. In this paper, we clarify the difference in the area of gaze region under different sound conditions; monaural (1.0 ch), stereo (2.0 ch), and surround (5.1 ch). We measured the gaze of subjects viewing video scenes of natural landscapes and calculated the horizontal and vertical standard deviations of the gaze points and the area of gaze region defined from them. A two-way analysis of variance shows the main effects of each of the sound and scene conditions on the vertical standard deviation of the gaze points and the interaction effect and the main effect of the sound condition on the area of gaze region. In addition, a multiple comparison test shows the significant differences in the vertical standard deviation for the interaction effect between the sound and scene conditions and in the area of gaze region for the sound conditions between monaural and stereo or surround. The hypothesis that the variance of gaze points is expanded by the surround sound is confirmed experimentally.

DOI： 10.11184/his.24.3_195

車両歩行者間のインタラクション行動のモデル化のための 2 段階入出力隠れマルコフモデル査読有り

新村文郷, 赤井直紀, 平山高嗣, 劉海龍, 川西康友, 出口大輔, 村瀬洋

情報処理学会論文誌 63 巻 ( 8 ) 頁： 1371 - 1382 2022年8月

記述言語：日本語掲載種別：研究論文（学術雑誌）

自動車と歩行者が交差するシーンに着目し，車両歩行者間のインタラクティブな行動をモデル化するための2段階入出力隠れマルコフモデル（TS-IOHMM）を提案する．従来の入出力隠れマルコフモデルは，相手の行動に依存して自己の意図と行動が決定される過程をモデル化した．提案するTS-IOHMMは，相手の行動の影響に加え，自己の行動もまた相手の意図に影響を与える点を考慮したモデルで，これまで注目されなかった意図と行動の変化にある法則性（ルール）の獲得を目指したモデルである．検証用に設定したルールに従って意図や行動が変化するシミュレーションデータを作成し，それを用いて提案モデルがどのようなルールを獲得できるか検証する実験を行った．その結果から，提案モデルがシミュレーションと同等の意図や行動を出力する行動ルールを獲得できたことを確認した．

DOI： 10.20729/00218987

Intuitive Gait Modeling using Mimetic-Words for Gait Description and Generation 査読有り

Hirotaka Kato, Takatsugu Hirayama, Keisuke Doman, Ichiro Ide, Yasutomo Kawanishi, Takahiro Komamizu, Daisuke Deguchi, Hiroshi Murase

頁： 240 - 245 2022年8月

Context-Aware Contribution Estimation for Feature Aggregation in Video Face Recognition 査読有り

Meng Zhang, Rujie Liu, Daisuke Deguchi, Hiroshi Murase

IEEE Access 10 巻頁： 79301 - 79310 2022年7月

記述言語：英語掲載種別：研究論文（学術雑誌）

The difficulties in video-based face recognition, such as dramatic pose variations and low quality, can be alleviated by leveraging the rich complementary information between the frames. However, limited by the mini-batch training strategy, the current deep learning methods only utilizes the frames in each batch during training, which ignore the context of the entire video. In this paper, we propose a context-aware feature aggregation scheme to aggregate complementary information between different frames. Firstly, a two-branch structure is designed as the C ontext-aware feature A ggregation N etwork ( CAN ). Secondly, a context-aware training strategy using a context bank is proposed, which alleviates the limitation of mini-batch samples by using the context of the entire video or several images belonging to the same ID and thus achieves global contribution estimation result. Comparative studies on benchmark datasets, such as IJB-C, YouTube Face (YTF), PaSC and COX, confirm that the proposed approach can achieve state-of-the-art level. Meanwhile, qualitative analysis on Multi-PIE dataset indicates that the contribution learned by the CAN is reasonable and beneficial to video face recognition.

DOI： 10.1109/ACCESS.2022.3193787

SfM-student: SfM法を用いたデータ拡張による列車前方映像からのセマンティックセグメンテーション査読有り

振津勇紀, 出口大輔, 川西康友, 井手一郎, 村瀬洋, 向嶋宏記, 長峯望

電子情報通信学会論文誌 J105-A 巻 ( 4 ) 頁： 48 - 57 2022年4月

記述言語：日本語掲載種別：研究論文（学術雑誌）

重要な公共交通機関として広く社会に普及している鉄道の沿線には，信号機や踏切など列車の安全運行を支える多くの設備が設置されている．これらの設備の日常的な整備や設置状況などに関する情報収集業務の多くは人手により行われており，その維持管理作業の自動化・効率化を実現する技術が強く求められている．このような課題に対して，営業運転中の列車に搭載したカメラにより前方を撮影した列車前方映像のみを用い，セマンティックセグメンテーションを施すことで鉄道環境における沿線設備などを自動認識する技術に期待が寄せられている．しかし，セマンティックセグメンテーションでは画素単位でクラス情報を人手で付与した学習データが必要であり，高い性能を得るために必要な大量の学習データを用意するコストは非常に高い．そこで本論文では，教師なしデータに対するセマンティックセグメンテーション結果とStructure from Motion（SfM）法による3次元復元結果を組み合わせることによってラベル付き3次元点群を生成し，それらを画像平面に投影することで擬似的なデータ拡張を行うSfM-student法を提案する．これにより，限られたラベルあり学習データのみからセマンティックセグメンテーションの精度向上を図る．実際の鉄道環境で撮影したデータを用いた実験を行ったところ，提案する3次元情報を利用した擬似的なデータ拡張手法は既存のデータ拡張手法と比べてセマンティックセグメンテーション精度を向上させることを確認した．

DOI： 10.14923/transfunj.2021JAP1021

歩行者属性と運転スキルを考慮したリスクポテンシャル関数による歩行者に対する安全マージン予測査読有り

前川大和, 平山高嗣, 川西康友, 出口大輔, 劉海龍, 井手一郎, 村瀬洋

電子情報通信学会論文誌 J105-A 巻 ( 3 ) 頁： 38 - 42 2022年3月

記述言語：日本語掲載種別：研究論文（学術雑誌）

本論文では，電動車いすのより安全な自動運転のための経路計画を目的として，歩行者属性と運転スキルを考慮したリスクポテンシャル関数を運転データから推定する手法を提案し，それらによる歩行者に対する電動車いすの安全マージンの違いを明らかにする．

DOI： 10.14923/transfunj.2021JAL2005

Masked face recognition with mask transfer and self-attention under the COVID-19 pandemic 査読有り

Zhang Meng, Rujie Liu, Daisuke Deguchi, Hiroshi Murase

IEEE Access 10 巻頁： 20527 - 20538 2022年2月

記述言語：英語掲載種別：研究論文（学術雑誌）

Face masks bring a new challenge to face recognition systems especially against the background of the COVID-19 pandemic. In this paper, a method mitigating the negative effects of mask defects on face recognition is proposed. Firstly, a low-cost, accurate method of masked face synthesis, i.e. mask transfer, is proposed for data augmentation. Secondly, an attention-aware masked face recognition (AMaskNet) is proposed to improve the performance of masked face recognition, which includes two modules: a feature extractor and a contribution estimator. Therein, the contribution estimator is employed to learn the contribution of the feature elements, thus achieving refined feature representation by simple matrix multiplications. Meanwhile, the end-to-end training strategy is utilized to optimize the entire model. Finally, a mask-aware similarity matching strategy(MS) is taken to improve the performance in the inference stage. The experiments show that the proposed method consistently outperforms on three masked face recognition datasets: RMFRD [1], COX [2] and Public-IvS [3]. Meanwhile, qualitative analysis experiments using CAM [4] indicate that the contribution learned by AMaskNet is more conducive to masked face recognition.

DOI： 10.1109/ACCESS.2022.3150345

Active Learning for Human Pose Estimation based on Temporal Pose Continuity 査読有り

Taro Mori, Daisuke Deguchi, Yasutomo Kawanishi, Ichiro Ide, Hiroshi Murase, Tetsuo Inoshita

Proceedings of International Workshop on Advanced Image Technology (IWAIT) 2022 2022年1月

Soft-boundary label relaxation with class placement constraints for semantic segmentation of the railway environment 査読有り

Yuki Furitsu, Daisuke Deguchi, Yasutomo Kawanishi, Ichiro Ide, Hiroshi Murase, Hiroki Mukojima, Nozomi Nagamine

Pattern Recognition Letters 150 巻頁： 258 - 264 2021年10月

Pointedness of an Image: Measuring How Pointy an Image is Perceived 査読有り

Chihaya Matsuhira, Marc A. Kastner, Ichiro Ide, Yasutomo Kawanishi, Takatsugu Hirayama, Keisuke Doman, Daisuke Deguchi, Hiroshi Murase

Proceedings of HCII 2021 37 巻頁： 137 - 144 2021年7月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

For computers to understand human perception, metrics that can capture human perception well are important. However, there are few metrics that characterize the visual perception of humans towards images. Therefore, in this paper, we propose a novel concept and a metric of pointedness of an image, which describes how pointy an image is perceived. The algorithm is inspired by the Features from Accelerated Segment Test (FAST) algorithm for corner detection which looks on the number of continuous neighboring darker pixels surrounding each pixel. We assume that this number would be proportional to the perceived pointedness in the region around the pixel. We evaluated our method towards how well it could capture the human perception of images. To compare the method with similar metrics that describe shapes, we prepared silhouette images of both artificial shapes and natural objects. The results showed that the proposed method gave nearly equivalent perceptual performance to other metrics and also worked in a larger variety of images.

DOI： https://doi.org/10.1007/978-3-030-78635-9_20

Best next-viewpoint recommendation by selecting minimum pose ambiguity for category-level object pose estimation 査読有り

Nik Mohd Zarifie Hashim, Yasutomo Kawanishi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

Journal of the Japan Society for Precision Engineering 87 巻 ( 5 ) 頁： 440 - 446 2021年5月

画像特徴点の対応を制約とした疎なLiDAR点群からのOdometry推定査読有り

清水政行, 川西康友, 出口大輔, 井手一郎, 村瀬洋

精密工学会誌 87 巻 ( 5 ) 頁： 447 - 454 2021年5月

光線方向の違いを考慮したライトフィールド特徴量の提案査読有り

清水政行, 川西康友, 出口大輔, 井手一郎, 村瀬洋

精密工学会誌 87 巻 ( 2 ) 頁： 197 - 205 2021年2月

記述言語：日本語掲載種別：研究論文（学術雑誌）

Many camera and photo applications, for example image mosaicing, Simultaneous Localization And Mapping (SLAM), and so on, use image keypoints, descriptors, and their matching. Recently, obtaining light-field has become easier thanks to the commercialization of light-field cameras. Light-field is a record of not only light ray intensity but also directions, and multi-view images can be acquired with it. In this paper, a light-field descriptor with light ray directions using light-field is proposed. A light-field descriptor is defined as a set of feature vectors that change according to the direction of the light ray. In addition, a method to match these descriptors is proposed. As the method focuses on the direction of the light ray, it allows more accurate matching than a conventional method when the camera pose changes against the target. To confirm the effectiveness, the proposed method is evaluated by using light-field image recorded in a real scene. As a result, the proposed method is shown to improve the accuraccy over a conventional method. In particular, the closer the distance between the target and the camera, the better accuracy of the proposed method could be achieved.

DOI： 10.2493/jjspe.87.197

ColAtt-Net: In Reducing the Ambiguity of Pedestrian Orientations on Attribute-Aware Semantic Segmentation Task 査読有り

Mahmud Dwi Sulistiyo, Yasutomo Kawanishi, Daisuke Deguchi, Ichiro Ide, Takatsugu Hirayama, Hiroshi Murase

IEEJ Transactions on Electrical and Electronic Engineering 16 巻 ( 2 ) 頁： 295 - 306 2021年2月

記述言語：英語掲載種別：研究論文（学術雑誌）

Semantic segmentation has become one of the trending topics in the world of computer vision and deep learning. Recently, due to an increasing demand to solve a semantic segmentation task simultaneously with attribute recognition of objects, a new task named attribute-aware semantic segmentation has been introduced. Since the task requires to handle pixel-wise object class estimation with its attributes such as a pedestrian’s body orientation, previous works had difficulties to handle ambiguous attributes such as body orientations in object-level, especially when segmenting the pedestrians with their attributes correctly. This paper proposes the ColAtt-Net that is an attribute-aware semantic segmentation model augmented by a column-wise mask branch to predict the pedestrians’ orientations in the horizontal perspective of the input image. We firmly assume that the pedestrians captured by a car-mounted camera are distributed horizontally so that for each column of the input image, the pedestrian pixels can be labeled with one orientation uniformly. In the proposed method, we split the output of the base semantic segmentation model into two branches; one branch for segmenting the object categories, while the other one, as the novel column-wise attribute branch, is to map the recognition of pedestrian’s orientations that are distributed horizontally. This method successfully enhances the performance of attribute-aware semantic segmentation by reducing the ambiguity on segmenting the pedestrian’s orientation. Improvements on the pedestrian orientation segmentation are confidently shown by the proposed method in the experimental results, both in quantitative and qualitative views. This paper also discusses how the improved performance becomes an advantage in the autonomous driving system.

DOI： https://doi.org/10.1002/tee.23296

LFIR2Pose: Pose Estimation from an Extremely Low-Resolution FIR Image Sequence 査読有り

Saki Iwata, Yasutomo Kawanishi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase, Tomoyoshi Aizawa

Proceedings of the 25th International Conference on Pattern Recognition 2021年1月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we propose a method for human pose estimation from a Low-resolution Far-InfraRed (LFIR) image sequence captured by a 16 × 16 FIR sensor array. Human body estimation from such a single LFIR image is a hard task. For training the estimation model, annotation of the human pose to the images is also a difficult task for human. Thus, we propose the LFIR2Pose model which accepts a sequence of LFIR images and outputs the human pose of the last frame, and also propose an automatic annotation system for the model training. Additionally, considering that the scale of human body motion is largely different among body parts, we also propose a loss function focusing on the difference. Through an experiment, we evaluated the human pose estimation accuracy using an original data set, and confirmed that human pose can be estimated accurately from an LFIR image sequence.

OMEga-GAN: Object Manifold Embedding GAN for Image Generation by Disentangling Parameters into Pose and Shape Manifolds 査読有り

Yasutomo Kawanishi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

Proceedings of the 25th International Conference on Pattern Recognition 頁： 7945 - 7952 2021年1月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we propose Object Manifold Embedding GAN (Ω-GAN) to generate images of variously shaped and arbitrarily posed objects from a noise variable sampled from a distribution defined over the pose and the shape manifolds in a vector space. We introduce Parametric Manifold Sampling to sample noise variables from a distribution over the pose manifold to conditionally generate object images in arbitrary poses by tuning the pose parameter. We also introduce Object Identity Loss for clearly disentangling the pose and shape parameters, which allows us to maintain the shape of the object instance when only the pose parameter is changed. Through evaluation, we confirmed that the proposed Ω-GAN could generate variously shaped object images in arbitrary poses by changing the pose and shape parameters independently. We also introduce an application of the proposed method for object pose estimation, through which we confirmed that the object poses in the generated images are accurate.

　前のページ - 次のページ　

<出口 大輔>

論文 - 出口 大輔

<出口　大輔>

論文 - 出口　大輔