Multi-attribute Residual Network (MAResNet) for Soft-Biometrics Recognition in Surveillance Scenarios

Bekele, Esubalew; Narber, Cody; Lawson, Wallace

doi:10.1109/fg.2017.55

Cited by 13 publications

(15 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The paper also proposes the challenging zeroshot identification scenario, where images of the target suspect are previously unseen by the learning algorithm. Since then, advances in deep learning CNNs have permeated re-id, necessary in tackling the high degrees of inter-class variation [10] when estimating body attributes [21], [22], [23], [24], [25], [26], [27], [28].…”

Section: Categorical Human Attribute Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Super-Fine Attributes with Crowd Prototyping

Martinho-Corbishley

Nixon

Carter

2019

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Recognising human attributes from surveillance footage is widely studied for attribute-based re-identification. However, most works assume coarse, expertly-defined categories, ineffective in describing challenging images. Such brittle representations are limited in descriminitive power and hamper the efficacy of learnt estimators. We aim to discover more relevant and precise subject descriptions, improving image retrieval and closing the semantic gap. Inspired by fine-grained and relative attributes, we introduce super-fine attributes, which now describe multiple, integral concepts of a single trait as multi-dimensional perceptual coordinates. Crowd prototyping facilitates efficient crowdsourcing of super-fine labels by pre-discovering salient perceptual concepts for prototype matching. We re-annotate gender, age and ethnicity traits from PETA, a highly diverse (19K instances, 8.7K identities) amalgamation of 10 re-id datasets including VIPER, CUHK and TownCentre. Employing joint attribute regression with the ResNet-152 CNN, we demonstrate substantially improved ranked retrieval performance with super-fine attributes in direct comparison to conventional binary labels, reporting up to a 11.2% and 14.8% mAP improvement for gender and age, further surpassed by ethnicity. We also find our 3 super-fine traits to outperform 35 binary attributes by 6.5% mAP for subject retrieval in a challenging zero-shot identification scenario.

show abstract

Section: Categorical Human Attribute Recognitionmentioning

confidence: 99%

“…The majority of attributes are extremely imbalanced, occurring in under 10% of the data, and do not include ethnicity, presumably due to its controversial nature. A number of works also experiment on PETA [21], [22], [23], [30], [53] for direct benchmarking.…”

Section: Approach and Dataset Overviewmentioning

confidence: 99%

Super-Fine Attributes with Crowd Prototyping

Martinho-Corbishley

Nixon

Carter

2019

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

“…Finally, attribute recognition is inherently a multi-label learning problem. Multiple person attributes may be present in an example image and there is inherent relationships among attributes [25], [2]. Three datasets were employed to demonstrate the effectiveness of deeper feature extraction pipeline compared to complicated branching and computationally expensive series of dense layers.…”

Section: A Datasetsmentioning

confidence: 99%

“…Following the evaluation metrics of [2], the performance metrics we employed are: mean average accuracy (mA), area under the curve (computed as averaged over all samples and attributes) (AUC), example based metrics of accuracy (Acc) (i.e., percentage where all attributes were correctly classified), precision (Prec), recall (Rec), and F1 score. These metrics display a balanced performance across the attributes.…”

Section: B Evaluation Metricsmentioning

confidence: 99%

The Deeper, the Better: Analysis of Person Attributes Recognition

Bekele

Lawson

2019

2019 14th IEEE International Conference on Automatic Face &Amp; Gesture Recognition (FG 2019)

Self Cite

View full text Add to dashboard Cite

In person attributes recognition, we describe a person in terms of their appearance. Typically, this includes a wide range of traits including age, gender, clothing, and footwear. Although this could be used in a wide variety of scenarios, it generally is applied to video surveillance, where attribute recognition is impacted by low resolution, and other issues such as variable pose, occlusion and shadow. Recent approaches have used deep convolutional neural networks (CNNs) to improve the accuracy in person attribute recognition. However, many of these networks are relatively shallow and it is unclear to what extent they use contextual cues to improve classification accuracy. In this paper, we propose deeper methods for person attribute recognition. Interpreting the reasons behind the classification is highly important, as it can provide insight into how the classifier is making decisions. Interpretation suggests that deeper networks generally take more contextual information into consideration, which helps improve classification accuracy and generalizability. We present experimental analysis and results for whole body attributes using the PA-100K and PETA datasets and facial attributes using the CelebA dataset.

show abstract

“…Studies on imbalanced data in deep learning have been conducted extensively [15]- [19]. In previous works [17]- [19], a weighted loss function was applied to a CNN in training to address the problem of unbalance data. Weight was defined in these studies using the number of samples of particular class.…”

Section: Drawbacks Of the Cnn With Heterogeneous Learningmentioning

confidence: 99%

Training of CNN with Heterogeneous Learning for Multiple Pedestrian Attributes Recognition Using Rarity Rate

Fukui

Yamashita

Yamauchi

et al. 2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Pedestrian attribute information is important function for an advanced driver assistance system (ADAS). Pedestrian attributes such as body pose, face orientation and open umbrella indicate the intended action or state of the pedestrian. Generally, this information is recognized using independent classifiers for each task. Performing all of these separate tasks is too time-consuming at the testing stage. In addition, the processing time increases with increasing number of tasks. To address this problem, multitask learning or heterogeneous learning is performed to train a single classifier to perform multiple tasks. In particular, heterogeneous learning is able to simultaneously train a classifier to perform regression and recognition tasks, which reduces both training and testing time. However, heterogeneous learning tends to result in a lower accuracy rate for classes with few training samples. In this paper, we propose a method to improve the performance of heterogeneous learning for such classes. We introduce a rarity rate based on the importance and class probability of each task. The appropriate rarity rate is assigned to each training sample. Thus, the samples in a mini-batch for training a deep convolutional neural network are augmented according to this rarity rate to focus on the classes with a few samples. Our heterogeneous learning approach with the rarity rate performs pedestrian attribute recognition better, especially for classes representing few training samples. key words: pedestrian attributes recognition, heterogeneous learning, rarity rate

show abstract

Multi-attribute Residual Network (MAResNet) for Soft-Biometrics Recognition in Surveillance Scenarios

Cited by 13 publications

References 22 publications

Super-Fine Attributes with Crowd Prototyping

Super-Fine Attributes with Crowd Prototyping

The Deeper, the Better: Analysis of Person Attributes Recognition

Training of CNN with Heterogeneous Learning for Multiple Pedestrian Attributes Recognition Using Rarity Rate

Contact Info

Product

Resources

About