3D-Aided Deep Pose-Invariant Face Recognition

Authors: Jian Zhao, Lin Xiong, Yu Cheng, Yi Cheng, Jianshu Li, Li Zhou, Yan Xu, Jayashree Karlekar, Sugiri Pranata, Shengmei Shen, Junliang Xing, Shuicheng Yan, Jiashi Feng

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Qualitative and quantitative experiments on both controlled and in-the-wild benchmarks clearly demonstrate superiority of the proposed model over state-of-the-arts.
Researcher Affiliation Collaboration 1National University of Singapore 2National University of Defense Technology 3Panasonic R&D Center Singapore 4National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 5Qihoo 360 AI Institute
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper states the network is implemented using "publicly available TensorFlow" but does not provide concrete access (a link or explicit statement of release) to the authors' own source code for the methodology described.
Open Datasets Yes For qualitative evaluation, we show visualized results on Multi-PIE [Gross et al., 2010] benchmark dataset. For quantitative evaluation, we evaluate face recognition performance on Multi-PIE and IJB-A datasets.
Dataset Splits Yes The images with 11 poses within 90 and 20 illumination levels of the first 150 identities are used for training. For testing, one frontal view with neutral expression and illumination (i.e., ID07) is used as the gallery image for each of the remaining 100 identities, and other images are used as probes. For training and testing, 10 random splits are provided by each protocol, respectively.
Hardware Specification Yes trained using Adam (β1=0.5) on a single NVIDIA TITAN X GPU with 12G memory.
Software Dependencies No The proposed network is implemented based on the publicly available Tensor Flow [Abadi et al., ] platform. The paper mentions TensorFlow but does not specify its version number or any other software dependencies with version numbers.
Experiment Setup Yes Throughout the experiments, the size of the RGB images of the input profile face (x), the simulator synthesis (x ), and the GLG prediction ( x) is fixed as 128 128; the sizes of the three RGB local patches (i.e., eyes, nose and mouth) are fixed as 80 40, 32 40, and 48 32, respectively; the dimensionality of the Gaussian random noise z is fixed as 100; the constraint factors λ0 to λ4 are empirically fixed as 0.05, 1.0, 0.1, 5 10 4, and 0.1, respectively; the batch size and learning rate are fixed as 16 and 3 10 5, respectively