Hiroshi TAKEDA

Training Robust Deep Neural Networks on Noisy Labels Using Adaptive Sample Selection With Disagreement

Hiroshi TAKEDA, Soh Yoshida, Mitsuji Muneyasu

Learning with noisy labels is one of the most practical but challenging tasks in deep learning. One promising way to treat noisy labels is to use the small-loss trick based on the memorization effect, that is, clean and noisy samples are identified …

jp (code pays fourni par la source)

Learning from Noisy Labeled Data Using Symmetric Cross-Entropy Loss for Image Classification

Hiroshi TAKEDA, Soh Yoshida, Mitsuji Muneyasu

While image classification with deep neural networks (DNNs) has made remarkable progress, learning from noisy labeled data degrades performance, and this problem remains challenging. Some previous methods, such as Co-teaching and MentorNet, use the memorization effects, which are characteristics of DNN training …

jp (code pays fourni par la source)

GENERATING SYNTHETIC TRAINING DATA FOR OBJECT DETECTION USING MULTI-TASK GENERATIVE ADVERSARIAL NETWORKS

Y.H. Lin, Katsuaki Suzuki, Hiroshi TAKEDA, Kazuaki Nakamura

Abstract. Nowadays, digitizing roadside objects, for instance traffic signs, is a necessary step for generating High Definition Maps (HD Map) which remains as an open challenge. Rapid development of deep learning technology using Convolutional Neural Networks (CNN) has achieved great success in …

jp (code pays fourni par la source)

CNN-based image recognition using mislabeled training data

Hiroshi TAKEDA, Soh Yoshida, Mitsuji Muneyasu

Tag-based Video Retrieval with Social Tag Relevance Learning

Hiroshi TAKEDA, Soh Yoshida, Mitsuji Muneyasu

High-quality tags play an important role in many applications such as multimedia information retrieval. This paper proposes a social tag relevance learning method using a data-driven approach to improving tag-based video retrieval performance. The tag relevance means how a tag is relevant …

jp (code pays fourni par la source)

FOREST COVER CLASSIFICATION USING GEOSPATIAL MULTIMODAL DATA

Katsuaki Suzuki, U. Rin, Yasunori Maeda, Hiroshi TAKEDA

Abstract. To address climate change, accurate and automated forest cover monitoring is crucial. In this study, we propose a Convolutional Neural Network (CNN) which mimics professional interpreters’ manual techniques. Using simultaneously acquired airborne images and LiDAR data, we attempt to reproduce the …

jp (code pays fourni par la source)

DRAWING FOR TRAFFIC MARKING USING BIDIRECTIONAL GRADIENT-BASED DETECTION WITH MMS LIDAR INTENSITY

Goro Takahashi, Hiroshi TAKEDA, Kentaro Nakamura

Abstract. Recently, the development of autonomous cars is accelerating on the integration of highly advanced artificial intelligence, which increases demand for a digital map with high accuracy. In particular, traffic markings are required to be precisely digitized since automatic driving utilizes them …

jp (code pays fourni par la source)

A FULLY AUTOMATED PIPELINE FOR CLASSIFICATION TASKS WITH AN APPLICATION TO REMOTE SENSING

Katsuaki Suzuki, Marc Claesen, Hiroshi TAKEDA, Bart De Moor

Abstract. Nowadays deep learning has been intensively in spotlight owing to its great victories at major competitions, which undeservedly pushed ‘shallow’ machine learning methods, relatively naive/handy algorithms commonly used by industrial engineers, to the background in spite of their facilities such as …

jp, be (code pays fourni par la source)

Methane production potential of subsurface microbes in Pleistocene sediments from a natural gas field of the dissolved-in-water type, central Japan

Hideyoshi Yoshioka, Hanako Mochimaru, Susumu Sakata, Hiroshi TAKEDA et autres

jp (code pays fourni par la source)

Automatic drawing for traffic marking with MMS LIDAR intensity

Goro Takahashi, Hiroshi TAKEDA, Y. Shimano

Abstract. Upgrading the database of CYBER JAPAN has been strategically promoted because the "Basic Act on Promotion of Utilization of Geographical Information", was enacted in May 2007. In particular, there is a high demand for road information that comprises a framework in …

jp (code pays fourni par la source)

Radioactive cesium behavior analysis in the refuse incineration plant (4) - simulation of the behavior of cesium in the incineration facility

Kaoru Yokoyama, Yoshiyuki Ohara, Noritake Sugitsue, Nobuo Takahashi et autres

Detection and Classification of Pole-like Objects from Mobile Laser Scanning Data of Urban Environments

H. Yokoyama, Hiroaka Date, Satoshi Kanai, Hiroshi TAKEDA

The Mobile Laser Scanning (MLS) system can acquire point clouds of urban environments including roads, buildings, trees, lamp posts etc. and enables effective mapping of them. With the spread of the MLS system, the demands for the management of roads and facilities …