A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion

Xiong, Lin; Jayashree, Karlekar; Zhao, Jian; Cheng, Yu-Ling; Xu, Yan; Feng, Jiashi; Pranata, Sugiri; Shen, Shengmei

doi:10.48550/arxiv.1704.00438

Cited by 10 publications

(7 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another strategy is to train the network for joint identification-verification task [29,32,33]. Xiong et al [36] proposed transferred deep feature fusion (TDFF) which infolves two-stage fusion of features trained with different networks and datasets. Template adaptation [8] is applied to further boost the performance.…”

Section: Related Workmentioning

confidence: 99%

L2-constrained Softmax Loss for Discriminative Face Verification

Ranjan¹,

Castillo²,

Chellappa³

2017

Preprint

126

280

View full text Add to dashboard Cite

In recent years, the performance of face verification systems has significantly improved using deep convolutional neural networks (DCNNs). A typical pipeline for face verification includes training a deep network for subject classification with softmax loss, using the penultimate layer output as the feature descriptor, and generating a cosine similarity score given a pair of face images. The softmax loss function does not optimize the features to have higher similarity score for positive pairs and lower similarity score for negative pairs, which leads to a performance gap. In this paper, we add an L 2 -constraint to the feature descriptors which restricts them to lie on a hypersphere of a fixed radius. This module can be easily implemented using existing deep learning frameworks. We show that integrating this simple step in the training pipeline significantly boosts the performance of face verification. Specifically, we achieve state-of-the-art results on the challenging IJB-A dataset, achieving True Accept Rate of 0.909 at False Accept Rate 0.0001 on the face verification protocol. Additionally, we achieve state-of-the-art performance on LFW dataset with an accuracy of 99.78%, and competing performance on YTF dataset with accuracy of 96.08%.

show abstract

Section: Related Workmentioning

confidence: 99%

L2-constrained Softmax Loss for Discriminative Face Verification

Ranjan¹,

Castillo²,

Chellappa³

2017

Preprint

126

280

View full text Add to dashboard Cite

show abstract

“…For example, Sankaranarayanan et al [17] proposed to retrain networks by using a triplet probability embedding (TPE) loss function and achieved good results on the IJB-A benchmark [10]. Xiong et al [29] proposed a framework named Transferred Deep Feature Fusion (TDFF) to fuse the features from two different networks trained on different datasets and learn a face classifiers in the target domain, which achieved state-of-the-art performance on IJB-A dataset. Mittal et al [15] developed a sketch-photo face matching system by fine-tuning the features of a stacked Auto-encoder and Deep Belief Network trained on a larger unconstrained face dataset.…”

Section: Deep Face Recognitionmentioning

confidence: 99%

DocFace: Matching ID Document Photos to Selfies

Shi

Jain

2018

2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems (BTAS)

View full text Add to dashboard Cite

Numerous activities in our daily life, including transactions, access to services and transportation, require us to verify who we are by showing our ID documents containing face images, e.g. passports and driver licenses. An automatic system for matching ID document photos to live face images in real time with high accuracy would speed up the verification process and remove the burden on human operators. In this paper, by employing the transfer learning technique, we propose a new method, DocFace, to train a domain-specific network for ID document photo matching without a large dataset. Compared with the baseline of applying existing methods for general face recognition to this problem, our method achieves considerable improvement. A cross validation on an ID-Selfie dataset shows that DocFace improves the TAR from 61.14% to 92.77% at FAR=0.1%. Experimental results also indicate that given more training data, a viable system for automatic ID document photo matching can be developed and deployed.

show abstract

“…We expect scholars and experts of different domains to seek out paradigms not thought of by us in the moment. Hence, whether it be an improved variant of adapting templates and feature fusion (e.g., like in [72]), deciding when to fuse, a new method of integration, along with the integration details, are open research questions. The data outweighs the benchmarks.…”

Section: Discussionmentioning

confidence: 99%

Families In Wild Multimedia: A Multimodal Database for Recognizing Kinship

Robinson,

Khan,

Yin

et al. 2020

Preprint

View full text Add to dashboard Cite

Recognizing kinship -a soft biometric with vast applications -in photos has piqued the interest of many machine vision researchers. The large-scale Families In the Wild (FIW) database promoted the problem by supporting annual kinshipbased vision challenges that saw consistent performance improvements. We have now begun to approach performance levels for image-based systems acceptable for practical use -something unforeseeable a decade ago. However, biometric systems can benefit from multi-modal perspectives, as information contained in multimedia can add to and complement that of still images. Thus, we aim to narrow the gap from research-to-reality by extending FIW with multimedia data (i.e., video, audio, and contextual transcripts). Specifically, we introduce the first large-scale dataset for recognizing kinship in multimedia, the FIW in Multimedia (FIW-MM) database. We utilize automated machinery to collect, annotate, and prepare the data with minimal human input and no financial cost. This large-scale, multimedia corpus allows problem formulations to follow more realistic template-based protocols. We show significant improvements in benchmarks for multiple kin-based tasks when additional media-types are added. Experiments provide insights by highlighting edge cases to inspire future research and areas of improvement. Emphasis is put on short and long-term research directions, with the overarching intent to increase the potential of systems built to automatically detect kinship in multimedia. Furthermore, we expect a broader range of researchers with recognition tasks, generative modeling, speech understanding, and nature-based narratives.

show abstract

A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion

Cited by 10 publications

References 30 publications

L2-constrained Softmax Loss for Discriminative Face Verification

L2-constrained Softmax Loss for Discriminative Face Verification

DocFace: Matching ID Document Photos to Selfies

Families In Wild Multimedia: A Multimodal Database for Recognizing Kinship

Contact Info

Product

Resources

About