Facial Attribute Capsules for Noise Face Super Resolution

Authors: Jingwei Xin, Nannan Wang, Xinrui Jiang, Jie Li, Xinbo Gao, Zhifeng Li12476-12483

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive benchmark experiments show that our method achieves superior hallucination results and outperforms state-of-the-art for very low resolution (LR) noise face image super resolution.
Researcher Affiliation Collaboration State Key Laboratory of Integrated Services Networks, School of Electronic Engineering, Xidian University, Xi an 710071, China State Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi an 710071, China Tencent AI Lab, China
Pseudocode No The paper describes the architecture and methods using text and diagrams (Figure 2, Figure 3), but does not contain any formal pseudocode or algorithm blocks.
Open Source Code No The paper states 'We implement our model using the pytorch environment' but does not provide any explicit statement of code release or a link to a repository.
Open Datasets Yes We conduct experiments on celeb A dataset (Liu et al. 2015).
Dataset Splits No The paper states 'We use the first 36000 images for training, and the following 1000 images for testing.' but does not mention a separate validation set split.
Hardware Specification Yes Training a basic FACN on celeb A dataset generally takes 10 hours with one Titan X Pascal GPU.
Software Dependencies No The paper states 'We implement our model using the pytorch environment,' but does not specify version numbers for PyTorch or any other software dependencies.
Experiment Setup Yes The momentum parameter is set to 0.5, weight decay is set to 1 10 4, and the initial learning rate is set to 3 10 4 and being divided a half every 20 epochs. The batch size is set to 16. We empirically set λ = 1, γP = 0.01 and γD = 0.01.