1 code implementation • 3 Sep 2023 • Onkar Krishna, Hiroki Ohashi, Saptarshi Sinha
A source sample is considered suitable if it differs from the target sample only in domain, without differences in unimportant characteristics such as orientation and color, which can hinder the model's focus on aligning the domain difference.
no code implementations • 7 Nov 2022 • Yuki Inoue, Hiroki Ohashi
However, as it is still in the early days of importing modular ideas to EIF, a search for modules effective in the EIF task is still far from a conclusion.
1 code implementation • 7 Sep 2022 • Saptarshi Sinha, Hiroki Ohashi
Long-tailed datasets, where head classes comprise much more training samples than tail classes, cause recognition models to get biased towards the head classes.
Ranked #10 on Long-tail Learning on Places-LT
1 code implementation • 29 Jul 2022 • Saptarshi Sinha, Hiroki Ohashi, Katsuyuki Nakamura
Further, we use the difficulty measures of each class to design a novel weighted loss technique called `class-wise difficulty based weighted (CDB-W) loss' and a novel data sampling technique called `class-wise difficulty based sampling (CDB-S)'.
no code implementations • 15 Nov 2021 • Hiroki Ohashi, Hiroto Nagayoshi
This study tackles on a new problem of estimating human-error potential on a shop floor on the basis of wearable sensors.
1 code implementation • 7 Sep 2021 • Katsuyuki Nakamura, Hiroki Ohashi, Mitsuhiro Okada
We compared the proposed sensor-fusion method with strong baselines on the MMAC Captions dataset and found that using sensor data as supplementary information to the egocentric-video data was beneficial, and that our proposed method outperformed the strong baselines, demonstrating the effectiveness of the proposed method.
1 code implementation • CVPR 2021 • Masato Tamura, Hiroki Ohashi, Tomoaki Yoshinaga
We propose a simple, intuitive yet powerful method for human-object interaction (HOI) detection.
Human-Object Interaction Concept Discovery Human-Object Interaction Detection
1 code implementation • ICLR 2021 • Naoyuki Terashita, Hiroki Ohashi, Yuichi Nonaka, Takashi Kanemaru
To this end, (1) we propose an influence estimation method that uses the Jacobian of the gradient of the generator's loss with respect to the discriminator's parameters (and vice versa) to trace how the absence of an instance in the discriminator's training affects the generator's parameters, and (2) we propose a novel evaluation scheme, in which we assess harmfulness of each training instance on the basis of how GAN evaluation metric (e. g., inception score) is expect to change due to the removal of the instance.
1 code implementation • 5 Oct 2020 • Saptarshi Sinha, Hiroki Ohashi, Katsuyuki Nakamura
We claim that the 'difficulty' of a class as perceived by the model is more important to determine the weighting.
Ranked #1 on Long-tail Learning on EGTEA
no code implementations • 2 Aug 2018 • Hiroki Ohashi, Mohammad Al-Naser, Sheraz Ahmed, Katsuyuki Nakamura, Takuto Sato, Andreas Dengel
ZSL classifies instances of unseen classes, from which no training data is available, by utilizing the attributes of the classes.