Zobrazeno 1 - 10
of 96
pro vyhledávání: '"Hirokatsu Kataoka"'
Publikováno v:
IEEE Access, Vol 12, Pp 142291-142305 (2024)
Pre-training for 3D object recognition typically requires a large-scale 3D dataset to learn effective 3D geometric representations. However, constructing such datasets is costly due to the extensive 3D data collection and human annotation required. T
Externí odkaz:
https://doaj.org/article/f49bdbe143ab42aabd174ce17d8fae06
Autor:
Risa Shinoda, Ko Motoki, Kensho Hara, Hirokatsu Kataoka, Ryohei Nakano, Tetsuya Nakazaki, Ryozo Noguchi
Publikováno v:
Smart Agricultural Technology, Vol 5, Iss , Pp 100271- (2023)
In cut-flower cultivation, production planning is an important task because demand fluctuates throughout the year. For precise cultivation planning, understanding the cultivation status is necessary by the growing stage. However, manually counting al
Externí odkaz:
https://doaj.org/article/273035c774174c39b9c70d4e2671cb17
Publikováno v:
Smart Agricultural Technology, Vol 4, Iss , Pp 100196- (2023)
With the recent development of computer vision technology, various computer vision techniques have been applied to agriculture. Recently, the Transformer network has been introduced to image recognition, which allows a different approach to extractin
Externí odkaz:
https://doaj.org/article/4e7e1fadcb78456b9a373749a2075f81
Publikováno v:
IEEE Access, Vol 11, Pp 136166-136178 (2023)
Does formula-driven supervised learning (FDSL) work effectively with fine-tuning on small datasets? Additionally, how many natural images do a network pre-trained with FDSL require to acquire sufficient image features? FDSL is a pre-training method t
Externí odkaz:
https://doaj.org/article/fd5b25a4ba32476391ddd53beaf9caa6
Publikováno v:
IEEE Access, Vol 11, Pp 120402-120418 (2023)
We present a novel framework for simulating time-varying decay effects for natural images. Conventional methods assume the input image includes enough decay information and uses the color or texture information of the decayed regions to transfer its
Externí odkaz:
https://doaj.org/article/07fff4e159f24c02842ba97fbbf79377
Publikováno v:
Sensors, Vol 23, Iss 13, p 6028 (2023)
In the field of embodied AI, vision-and-language navigation (VLN) is a crucial and challenging multi-modal task. Specifically, outdoor VLN involves an agent navigating within a graph-based environment, while simultaneously interpreting information fr
Externí odkaz:
https://doaj.org/article/cb588ffb3aa94904800a70fdabc7f57d
Publikováno v:
Applied Sciences, Vol 11, Iss 14, p 6287 (2021)
Summaries of scientific publications enable readers to gain an overview of a large number of studies, but users’ preferences have not yet been explored. In this paper, we conduct two user studies (i.e., short- and long-term studies) where Japanese
Externí odkaz:
https://doaj.org/article/2e417c555efd46258b7bd2721db1d6ae
Publikováno v:
Sensors, Vol 20, Iss 19, p 5647 (2020)
Virtual Try-on is the ability to realistically superimpose clothing onto a target person. Due to its importance to the multi-billion dollar e-commerce industry, the problem has received significant attention in recent years. To date, most virtual try
Externí odkaz:
https://doaj.org/article/39dec57f4e78495a87bc8f6356bccdbb
Publikováno v:
Sensors, Vol 20, Iss 17, p 4761 (2020)
This study proposes a framework for describing a scene change using natural language text based on indoor scene observations conducted before and after a scene change. The recognition of scene changes plays an essential role in a variety of real-worl
Externí odkaz:
https://doaj.org/article/3169401acabd47dca1f4aeba71180bb4
Publikováno v:
Sensors, Vol 20, Iss 8, p 2281 (2020)
This paper proposes a framework that allows the observation of a scene iteratively to answer a given question about the scene. Conventional visual question answering (VQA) methods are designed to answer given questions based on single-view images. Ho
Externí odkaz:
https://doaj.org/article/d85a4ca10a724f53b6a8aa1df363e7fd