Proceedings ArticleDOI
Matching cross-resolution face images using co-transfer learning
Himanshu Bhatt,Richa Singh,Mayank Vatsa,Nalini K. Ratha +3 more
- pp 1453-1456
Reads0
Chats0
TLDR
A co-transfer learning framework is proposed in which knowledge learnt in controlled high resolution environment is transferred for matching low resolution probe images with high resolution gallery and the proposed algorithm outperforms existing approaches by at least 5%.Abstract:
Face recognition systems, trained in controlled environment, often fail to efficiently match low resolution images with high resolution images. In this research, a co-transfer learning framework is proposed in which knowledge learnt in controlled high resolution environment is transferred for matching low resolution probe images with high resolution gallery. The proposed framework seamlessly combines transfer learning and co-training to perform knowledge transfer by updating classifier's decision boundary with low resolution probe instances. Experiments are performed on the CMU-Multi-PIE and SCface database with gallery images of size 72 × 72 and size of probe images varying from 48 × 48 to 16 × 16. The results show that, in terms of rank-1 identification accuracy, the proposed algorithm outperforms existing approaches by at least 5%.read more
Citations
More filters
Journal ArticleDOI
Online learning: A comprehensive survey
TL;DR: Online learning as mentioned in this paper is a family of machine learning methods, where a learner attempts to tackle some predictive (or any type of decision-making) task by learning from a sequence of data instances one by one at each time.
Journal ArticleDOI
Improving cross-resolution face matching using ensemble-based co-transfer learning.
TL;DR: A co-transfer learning framework is proposed, which is a cross-pollination of transfer learning and co-training paradigms and is applied for cross-resolution face matching and enhances the performance of cross- resolution face recognition.
Dissertation
Pushing boundaries of face recognition : adversary, heterogeneity, and scale
TL;DR: This dissertation addresses the challenge of heterogeneous face matching scenarios, such as matching a sketch against a mugshot dataset of digital photographs, cross-spectrum, and crossresolution matching, that arise in a wide range of law enforcement scenarios, and develops an approach to efficiently update the face recognition engine to incorporate incremental training data.
Journal ArticleDOI
SUPREAR-NET: Supervised Resolution Enhancement and Recognition Network
TL;DR: A Supervised Resolution Enhancement and Recognition Network (SUPREAR-NET), which does not corrupt the useful class-specific information of the face image and transforms a low resolution probe image into a high resolution one, followed by effective matching with the gallery using a trained discriminative model.
Proceedings ArticleDOI
LC-DECAL: Label Consistent Deep Collaborative Learning for Face Recognition
TL;DR: The proposed Label Consistent Deep Collaborative Learning (LC-DECAL) framework makes use of label consistency, transfer learning, ensemble learning, and co-training for training a deep neural network for the target domain.
References
More filters
Journal ArticleDOI
Distinctive Image Features from Scale-Invariant Keypoints
TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Proceedings ArticleDOI
Combining labeled and unlabeled data with co-training
Avrim Blum,Tom M. Mitchell +1 more
TL;DR: A PAC-style analysis is provided for a problem setting motivated by the task of learning to classify web pages, in which the description of each example can be partitioned into two distinct views, to allow inexpensive unlabeled data to augment, a much smaller set of labeled examples.
Journal ArticleDOI
Image Super-Resolution Via Sparse Representation
TL;DR: This paper presents a new approach to single-image superresolution, based upon sparse signal representation, which generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods.
Journal ArticleDOI
Single-Image Super-Resolution Using Sparse Regression and Natural Image Prior
Kwang In Kim,Younghee Kwon +1 more
TL;DR: Compared with existing algorithms, KRR leads to a better generalization than simply storing the examples as has been done in existing example-based algorithms and results in much less noisy images.
ReportDOI
Support Vector Machines Applied to Face Recognition
TL;DR: A SVM -based face recognition algorithm that is compared with a principal component analysis (PCA) based algorithm on a difficult set of images from the FERET database and generated a similarity metric between faces that is learned from examples of differences between faces.