Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks

2016/08/26 by Kaipeng Zhang, Zhanpeng Zhang, Zhifeng Li +1 · 44 citations
Computer Science · #Biometric Identification and Security #Face and Expression Recognition #Face recognition and analysis

paper · doi:10.1109/lsp.2016.2603342

openalex created_date 2016/06/24 · openalex publication_date 2016/08/26 · crossref created 2016/08/26 · crossref issued 2016/10/01 · crossref published 2016/10/01 · crossref published-print 2016/10/01 · crossref deposited 2022/01/12 · crossref indexed 2026/07/30 · openalex updated_date 2026/08/01

Abstract

Face detection and alignment in unconstrained environment are challenging due to various poses, illuminations, and occlusions. Recent studies show that deep learning approaches can achieve impressive performance on these two tasks. In this letter, we propose a deep cascaded multitask framework that exploits the inherent correlation between detection and alignment to boost up their performance. In particular, our framework leverages a cascaded architecture with three stages of carefully designed deep convolutional networks to predict face and landmark location in a coarse-to-fine manner. In addition, we propose a new online hard sample mining strategy that further improves the performance in practice. Our method achieves superior accuracy over the state-of-the-art techniques on the challenging face detection dataset and benchmark and WIDER FACE benchmarks for face detection, and annotated facial landmarks in the wild benchmark for face alignment, while keeps real-time performance.

Citations

• Robust Real-Time Face Detection

Cited by

• Making Machine Learning Accessible for Developmental Science: The Case of Automated Face Detection
• Beyond Inference Intervention: Identity-Decoupled Diffusion for Face Anonymization
• Super-recognizers sample visual information of superior computational value for facial recognition
• Generating Auxiliary Tasks with Reinforcement Learning
• Cross-Enhanced Multimodal Fusion of Eye-Tracking and Facial Features for Alzheimer's Disease Diagnosis
• Investigating Adversarial Robustness against Preprocessing used in Blackbox Face Recognition
• Inference-Time Search using Side Information for Diffusion-based Image Reconstruction
• Conformal Inference for Open-Set and Imbalanced Classification
• ROFI: A Deep Learning-Based Ophthalmic Sign-Preserving and Reversible Patient Face Anonymizer
• BiFuseNet: A Multimodal Network for Estimating Blood Alcohol Concentration via Bidirectional Hierarchical Fusion
• CL-UZH submission to the NIST SRE 2024 Speaker Recognition Evaluation
• DGM4+: Dataset Extension for Global Scene Inconsistency
• Analysis of Bias in Deep Learning Facial Beauty Regressors
• Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention
• Incorporating Scene Context and Semantic Labels for Enhanced Group-level Emotion Recognition
• CusEnhancer: A Zero-Shot Scene and Controllability Enhancement Method for Photo Customization via ResInversion
• CURE: Centroid-guided Unsupervised Representation Erasure for Facial Recognition Systems
• Foundation Models for Face Presentation Attack Detection: A Unified Linear-Probing Benchmark
• CPCLDETECTOR: Knowledge Enhancement and Alignment Selection for Chinese Patronizing and Condescending Language Detection
• Controllable Localized Face Anonymization Via Diffusion Inpainting
• Hierarchical MLANet: Multi-level Attention for 3D Face Reconstruction From Single Images
• Spoken in Jest, Detected in Earnest: A Systematic Review of Sarcasm Recognition -- Multimodal Fusion, Challenges, and Future Prospects
• Human Preference-Aligned Concept Customization Benchmark via Decomposed Evaluation
• Uncertainty-Aware Collaborative System of Large and Small Models for Multimodal Sentiment Analysis
• Paired-Sampling Contrastive Framework for Joint Physical-Digital Face Attack Detection
• Taming Transformer for Emotion-Controllable Talking Face Generation
• ViT-FIQA: Assessing Face Image Quality using Vision Transformers
• Unified Knowledge Distillation Framework: Fine-Grained Alignment and Geometric Relationship Preservation for Deep Face Recognition
• A Multimodal Neural Network for Recognizing Subjective Self-Disclosure Towards Social Robots
• Forgery Guided Learning Strategy with Dual Perception Network for Deepfake Cross-domain Detection
• Ensembling Synchronisation-based and Face-Voice Association Paradigms for Robust Active Speaker Detection in Egocentric Recordings
• FaceAnonyMixer: Cancelable Faces via Identity Consistent Latent Space Mixing
• Disentangling Bias by Modeling Intra- and Inter-modal Causal Attention for Multimodal Sentiment Analysis
• Backdoor Attacks on Deep Learning Face Detection
• Latent Diffusion Based Face Enhancement under Degraded Conditions for Forensic Face Recognition
• Robust Deepfake Detection for Electronic Know Your Customer Systems Using Registered Images
• Gems: Group Emotion Profiling Through Multimodal Situational Understanding
• Investigation of Accuracy and Bias in Face Recognition Trained with Synthetic Data
• In-context Learning of Vision Language Models for Detection of Physical and Digital Attacks against Face Recognition Systems
• Color histogram equalization and fine-tuning to improve expression recognition of (partially occluded) faces on sign language datasets
• Facial Demorphing from a Single Morph Using a Latent Conditional GAN
• Towards Trustworthy AI: Secure Deepfake Detection using CNNs and Zero-Knowledge Proofs
• InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing
• Approximate Processing Element Design and Analysis for the Implementation of CNN Accelerators