Rotating your face using multi-task deep neural network

By Junho Yim, Heechul Jung, ByungIn Yoo, Changkyu Choi, Du-sik Park, Junmo Kim. 2015

297 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

An Overview of Multi-task Learning (2018, background): “Specifically, applications of MTL based on facial images include face verification [84], personalized age estimation [85], multi-cue face recognition [86], head-pose estimation [22,87], facial landmark detection [18], and facial image rotation [88].”
Coupled Generative Adversarial Networks (2016, background): “Our work is also related to the prior works in cross-domain image generation [31, 32, 33], which studied transforming an image in one style to the corresponding images in another style.”
Cross-Stitch Networks for Multi-task Learning (2016, background): “earning and selection [4,13,31,37], transfer learning [39,41,56] etc. and their widespread applications in other ﬁelds, such as genomics [38], natural language processing [7,8,35] and computer vision [3,10,30,31,40,51,53,58]. In fact, many times multi-task learning is implicitly used without referen”
Dynamic Filter Networks (2016, background): “Yim et al. [25] and Yang et al. [24] learn to rotate a given face to another pose.”
A Survey on Multi-Task Learning (2017, background): “In [178], a multi-task deep model is proposed to rotate facial images to a”
Deep multi-task learning with low level tasks supervised at lower layers (2016, background): “Multiple-tasks supervision at the outermost layer has a strong tradition in neural net models in vision and elsewhere (Caruana, 1993; Zhang and Zhang, 2014; Yim et al., 2015).”
Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis (2017, methodology): “CPF [38] - - 71.”
Disentangled Representation Learning GAN for Pose-Invariant Face Recognition (2017, methodology): “L2 Loss Prior work [38, 40, 44] on face rotation normally employ the L2 loss to learn a mapping between two views.”
FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis (2018, background): “We would also like to acknowledge their contributions [12, 39, 40, 32].”
Model Compression with Two-stage Multi-teacher Knowledge Distillation for Web Question Answering System (2019, background): “mble of teachers. 2.2 Multi-task Learning Multi-task learning has been widely studied in deep learning, which leverages the information among different tasks to improve the generalization performance [3, 6, 28]. Fares et al. [7] empirically evaluated the utility of transfer and multi-task learning o”
A Light CNN for Deep Face Representation With Noisy Labels (2015, methodology): “our Light CNNs are potentially robust for agevariant problems. We also compare the Light CNNs with multi-view face recognition methods [67], [66], [50], [68] and pose-aware face image synthesis [70], [69] in Table VIII(b). It is obvious that the Light CNN-29 obtains great performance on Multi-PIE, w”
DeepNAT: Deep Convolutional Neural Network for Segmenting Neuroanatomy (2017, background): “While there have been applications for deep multi-task learning (Yim et al., 2015; Luong et al., 2015), we are not aware of previous applications for image segmentation.”
Pose Guided Person Image Generation (2017, background): “Researchers [34, 33, 11, 8] addressed the task of face image generation conditioned on a reference image and a specific face viewpoint.”
NDDR-CNN: Layerwise Feature Fusing in Multi-Task CNNs by Neural Discriminative Dimensionality Reduction (2018, unknown): “used face alignment and reconstruction as auxiliary tasks for face recognition [56].”
A Comprehensive Survey on Pose-Invariant Face Recognition (2015, methodology): “Zhu et al. [2013] and Yim et al. [2015] adopted CNN to extract pose-robust features, and face synthesis is accomplished via a full-connected reconstruction layer.”
Object Detection With Deep Learning: A Review (2018, background): “multiple correlated tasks from the same input [102], [103].”
Learning Pose-Aware Models for Pose-Invariant Face Recognition in the Wild (2019, methodology): “Contrary to methods that rely on multi-task learning [17] or [18] and model identity and viewpoint with a single network, we treat each type of alignment and data independently .”
Deep Face Recognition: A Survey (2018, background): “many to one recovering the canonical view of face images from one or many images of nonfrontal view SAE [80], [218], [195] CNN [230], [232], [73], [35], [201] GAN [75], [156], [37], [204]”
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving (2016, background): “An important application for multi-task learning is face recognition [60], [56], [44].”
Generative Adversarial Networks for Face Generation: A Survey (2022, methodology): “Disentanglement representation learning (DRL) was previously applied in multi-task CNNs [148, 157, 169] using joint models to learn feature transformations in the target subdomains.”