Looking beyond appearances: Synthetic training data for deep cnns in re-identification

IB Barbosa, M Cristani, B Caputo… - Computer Vision and …, 2018 - Elsevier
Re-identification is generally carried out by encoding the appearance of a subject in terms of
outfit, suggesting scenarios where people do not change their attire. In this paper we …

Learning semantic text features for web text-aided image classification

D Wang, K Mao - IEEE Transactions on Multimedia, 2019 - ieeexplore.ieee.org
The good generalization performance of conventional pattern classifiers often relies on the
size of training data labeled by costly human labor. These days, publicly available web …

Visual ground truth construction as faceted classification

F Giunchiglia, M Bagchi, X Diao - arXiv preprint arXiv:2202.08512, 2022 - arxiv.org
Recent work in Machine Learning and Computer Vision has provided evidence of
systematic design flaws in the development of major object recognition benchmark datasets …

Learning deep visual object models from noisy web data: How to make it work

N Massouh, F Babiloni, T Tommasi… - 2017 IEEE/RSJ …, 2017 - ieeexplore.ieee.org
Deep networks thrive when trained on large scale data collections. This has given ImageNet
a central role in the development of deep architectures for visual object classification …

Construction of diverse image datasets from web collections with limited labeling

NC Mithun, R Panda… - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
Image datasets play a pivotal role in advancing computer vision and multimedia research.
However, most of the datasets are created by extensive human effort and are extremely …

Generating diverse image datasets with limited labeling

NC Mithun, R Panda, AK Roy-Chowdhury - Proceedings of the 24th ACM …, 2016 - dl.acm.org
Image datasets play a pivotal role in advancing multimedia and image analysis research.
However, most of these datasets are created by extensive human effort and extremely …

[PDF][PDF] Creating training datasets for ocr in mobile device video stream.

DA Ilin, VE Krivtsov - ECMS, 2015 - scs-europe.net
This paper studies methods of data sampling for training of convolutional neural networks for
character recognition. These methods are considered for optical character recognition of …

Count on me: learning to count on a single image

F Setti, D Conigliaro, M Tobanelli… - IEEE Transactions on …, 2017 - ieeexplore.ieee.org
Individuating and locating repetitive patterns in still images is a fundamental task in image
processing, typically achieved by means of correlation strategies. In this paper, we provide a …

Long-term behavior understanding based on the expert-based combination of short-term observations in high-resolution CCTV

K Schutte, G Burghouts, N van der Stap… - … and Photonics for …, 2016 - spiedigitallibrary.org
The bottleneck in situation awareness is no longer in the sensing domain but rather in the
data interpretation domain, since the number of sensors is rapidly increasing and it is not …

RoboCup@ Home-Objects: benchmarking object recognition for home robots

N Massouh, L Brigato, L Iocchi - RoboCup 2019: Robot World Cup XXIII 23, 2019 - Springer
This paper presents a benchmark for object recognition inspired by RoboCup@ Home
competition and thus focusing on home robots. The benchmark includes a large-scale …