Use of Affine Invariants in Locally Likely Arrangement Hashing for Camera-Based Document Image Retrieval
- T. Nakai, K. Kise, M. Iwamura
- Computer ScienceInternational Workshop on Document Analysis…
- 13 February 2006
This paper introduces into LLAH an affine invariant instead of the perspective invariant so as to improve its adjustability and experimental results show that the use of the affines enables us to improve either the accuracy from 96.2% to 97.8%, or the retrieval time from 112 msec./query to 75 msec./ query by selecting parameters of processing.
Shakedrop Regularization for Deep Residual Learning
This paper proposes a new regularization method called ShakeDrop regularization, inspired by Shake-Shake, and introduces a training stabilizer, which is an unusual use of an existing regularizer.
Segmentation of Page Images Using the Area Voronoi Diagram
It is confirmed that the proposed method of page segmentation based on the approximated area Voronoi diagram is effective for extraction of body text regions, and it is as efficient as other methods based on connected component analysis.
Knowledge-Based and Intelligent Information and Engineering Systems
- A. Koenig, A. Dengel, Knut Hinkelmann, K. Kise, R. Howlett, L. Jain
- Computer ScienceLecture Notes in Computer Science
- 9 November 2011
The four-volume set LNAI 6881-LNAI 6884 constitutes the refereed proceedings of the 15th International Conference on Knowledge-Based Intelligent Information and Engineering Systems, KES 2011, held in…
- Yoshihiro Yamada, M. Iwamura, K. Kise
- Computer ScienceInternational Conference on Learning…
- 15 February 2018
ShakeDrop is inspired by Shake-Shake regularization that decreases error rates by disturbing learning and can be applied to not only ResNeXt but also ResNet, Wide ResNet and PyramidNet in a memory efficient way.
In the blink of an eye: combining head motion and eye blink frequency for activity recognition with Google Glass
- Shoya Ishimaru, K. Kunze, A. Bulling
- Computer Science, PsychologyInternational Conference on Adaptive Hypermedia…
- 7 March 2014
It is shown that eye blink frequency data from an unobtrusive, commercial platform which is not a dedicated eye tracker is good enough to be useful and that adding head motion patterns information significantly improves the recognition rates.
Page Segmentation Techniques in Document Analysis
- K. Kise
- Computer ScienceHandbook of Document Image Processing and…
Speech balloon and speaker association for comics and manga understanding
- Christophe Rigaud, N. L. Thanh, K. Kise
- Computer ScienceIEEE International Conference on Document…
- 23 August 2015
A speech balloon and comic character association method able to retrieve which character is emitting which speech balloon is proposed based on geometric graph analysis and anchor point selection is proposed.
Camera Based Document Image Retrieval with More Time and Memory Efficient LLAH
Improvements of the camera-based document image retrieval method with Locally Likely Arrangement Hashing with reduction of the required amount of memory by about 80% and that of processing time by about 60%.
I know what you are reading: recognition of document types using mobile eye tracking
- K. Kunze, Yuzuko Utsumi, Yuki Shiga, K. Kise, A. Bulling
- Computer ScienceInternational Semantic Web Conference
- 8 September 2013
This work investigates whether different document types can be automatically detected from visual behaviour recorded using a mobile eye tracker, and presents an initial recognition approach that uses special purpose eye movement features as well as machine learning for document type detection.