MinNet: Minutia Patch Embedding Network for Automated Latent Fingerprint Recognition

Öztürk, Halil İbrahim; Selbes, Berkay; Artan, Yusuf

doi:10.1109/cvprw56347.2022.00169

Cited by 8 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, from observation, we find that the differences in the pedestrian walking speed and camera frame rate have resulted in inconsistent frame length of the gait cycle (as shown in Figure 1), and thus the single temporal modeling approach cannot adapt to the diversity of motion. Furthermore, unlike face recognition [6,7], fingerprint recognition [8,9], etc., which extracts identity information from a single image, gait recognition technology is based on video sequences. In the early days, most methods [10][11][12] fused sequence features by generating template images.…”

Section: Introductionmentioning

confidence: 99%

Metric-Based Key Frame Extraction for Gait Recognition

Wei

Zhao

et al. 2022

Electronics

View full text Add to dashboard Cite

Gait recognition is one of the most promising biometric technologies that can identify individuals at a long distance. From observation, we find that there are differences in the length of the gait cycle and the quality of each frame in the sequence. In this paper, we propose a novel gait recognition framework to analyze human gait. On the one hand, we designed the Multi-scale Temporal Aggregation (MTA) module that models temporal and aggregate contextual information with different scales, on the other hand, we introduce the Metric-based Frame Attention Mechanism (MFAM) to re-weight each frame by the importance score, which calculates using the distance between frame-level features and sequence-level features. We evaluate our model on two of the most popular public datasets, CASIA-B and OU-MVLP. For normal walking, the rank-1 accuracies on the two datasets are 97.6% and 90.1%, respectively. In complex scenarios, the proposed method achieves accuracies of 94.8% and 84.9% on CASIA-B under bag-carrying and coat-wearing walking conditions. The results show that our method achieves the top level among state-of-the-art methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Metric-Based Key Frame Extraction for Gait Recognition

Wei

Zhao

et al. 2022

Electronics

View full text Add to dashboard Cite

show abstract

“…To thoroughly val-idate the effectiveness of the descriptors, we do not employ any preprocessing of the fingerprint images like image enhancement. The experiments demonstrate that our method outperforms other deep-learning based descriptor methods [3,32], conventional well-designed descriptor [5], and Commercial Off-The-Shelf (COTS) method [31]. Besides, DMD maintains good performance even after binarization, thus indicating its potential for practical applications as an automated fingerprint recognition system.…”

Section: Introductionmentioning

confidence: 78%

“…We retain the entire pipeline of their system, which includes image enhancement, the extraction of minutiae and virtual minutiae templates, as well as the processing of descriptors. Regarding MinNet [32], we train it with original patch fingerprint images instead of enhancement ones which are the same as ours, and increase the dimensionality of its descriptors to 768 to align with our configuration. As to commercial matcher VeriFinger v12.0, it has two types of matchers: ISO minutia-only template and proprietary template consisting of minutiae and other features.…”

Section: Compared Methodsmentioning

confidence: 99%

“…In contrast to the approach of directly selecting minutiae according to MCC [5] minutiae pair matching scores as presented in [32], our methodology incorporates a multitude of selection strategies. These strategies are designed to identify minutiae that not only are correctly matched but also resilient to distortion.…”

Section: Training Sample Generationmentioning

confidence: 99%

“…This approach, inherently resistant to overall fingerprint pose changes, provides superior accuracy to most fixedlength methods, albeit less efficiently [19]. Works by [5,30,32] have concentrated on encoding minutia relationships within a patch, designating one minutia as the anchor point, with Öztürk et al [32] incorporating CNNs for concurrent texture feature encoding. Beyond using minutiae as anchors, Cao et al [2,3] integrated orientation fields as dense anchors for comprehensive depictions, termed virtual minutiae or texture templates [1].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

How higher educational institutions cope with social change : the case of Tsinghua University, China

Pan¹,

潘甦燕.²

View full text Add to dashboard Cite

Recently, several methods have been proposed to estimate 3D human pose from multi-view images and achieved impressive performance on public datasets collected in relatively easy scenarios. However, there are limited approaches for extracting 3D human skeletons from multimodal inputs (e.g., RGB and pointcloud) that can enhance the accuracy of predicting 3D poses in challenging situations. We fill this gap by introducing a pipeline called PointVoxel that fuses multi-view RGB and pointcloud inputs to obtain 3D human poses. We demonstrate that volumetric representation is an effective architecture for integrating these different modalities. Moreover, in order to overcome the challenges of annotating 3D human pose labels in difficult scenarios, we develop a synthetic dataset generator for pretraining and design an unsupervised domain adaptation strategy so that we can obtain a well-trained 3D human pose estimator without using any manual annotations. We evaluate our approach on four datasets (two public datasets, one synthetic dataset, and one challenging dataset named BasketBall collected by ourselves), showing promising results. The code and dataset will be released soon.

show abstract