r-BTN: Cross-Domain Face Composite and Synthesis From Limited Facial Patches

Authors: Yang Song, Zhifei Zhang, Hairong Qi

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments have been conducted to demonstrate the superior performance from r BTN as compared to existing potential solutions.
Researcher Affiliation Academia Yang Song, Zhifei Zhang, Hairong Qi Department of Electrical Engineering and Computer Science University of Tennessee, Knoxville, TN 37996, USA {ysong18, zzhang61, hqi@utk.edu}
Pseudocode No The paper describes the algorithm using textual descriptions and mathematical equations along with flowcharts, but does not include a structured pseudocode or algorithm block.
Open Source Code No The paper mentions 'Details are shown in supplementary materials' but does not include an explicit statement about the release of its source code or a link to a code repository.
Open Datasets Yes We collect 1,577 face/sketch pairs from the datasets CUHK (Wang and Tang 2009), CUFSF (Zhang, Wang, and Tang 2011), AR (Martinez and Benavente 2007), FERET (Phillips et al. 2000), and IIIT-D (Bhatt et al. 2012). ... We collect frontal face images with uniform background and controlled illumination from datasets CFD (Ma, Correll, and Wittenbrink 2015), Siblings DB (Vieira et al. 2014), and PUT (Kasinski, Florek, and Schmidt 2008)...
Dataset Splits No The paper mentions 3,126 face/sketch pairs and 300 pairs for testing, but does not provide explicit details about training/validation/test splits beyond that, such as percentages or specific counts for all splits.
Hardware Specification No The paper describes implementation details but does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper mentions the use of 'ADAM (Kingma and Ba 2014)' but does not list specific software dependencies with version numbers (e.g., programming languages, deep learning frameworks, or libraries).
Experiment Setup Yes In the training, we adopt ADAM (Kingma and Ba 2014) (α = 0.0002, β = 0.5). ... The parameter λ in Eq. 3 is set to be 100. Details are shown in supplementary materials. After 100 epochs, we could achieve the results as shown in this paper.