Coupled CycleGAN: Unsupervised Hashing Network for Cross-Modal Retrieval

Authors: Chao Li, Cheng Deng, Lei Wang, De Xie, Xianglong Liu176-183

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on three popular benchmark datasets show that the proposed UCH outperforms the state-of-the-art unsupervised cross-modal hashing methods.
Researcher Affiliation Academia 1School of Electronic Engineering, Xidian University, Xi an 710071, China 2State Key Lab of Software Development Environment, Beihang University, Beijing 100191, China
Pseudocode Yes Algorithm 1 Optimizing process of the proposed UCH
Open Source Code No The paper does not provide any specific links to source code or explicitly state that the code for their methodology is made open-source or available.
Open Datasets Yes Three popular benchmark datasets in cross-modal retrieval: MIRFlickr-25K (Huiskes and Lew 2008), NUSWIDE (Chua et al. 2009), and Microsoft COCO (Lin et al. 2014) are adopted to validate our proposed method.
Dataset Splits No The paper describes 'training set' and 'retrieval set' for different datasets (e.g., 'For supervised baselines, 5,000 image-text pairs are selected from retrieval set to construct training set.'), but does not explicitly provide details about a distinct 'validation' dataset split used for hyperparameter tuning or early stopping.
Hardware Specification Yes UCH is implemented via Tensor Flow and executed on a server with two NVIDIA TITAN X GPUs.
Software Dependencies No The paper mentions 'Tensor Flow' as the implementation framework, but does not provide specific version numbers for TensorFlow or any other software libraries or dependencies.
Experiment Setup Yes In all our experiments, the initial leaning rates of image and text networks are set to 10 4 and 10 2. And batchsize and weight decay are set to 128 and 10 1. ... GI T f and GT I f are constructed with two different deep networks with four full-connected layers, e.g., (GI T f :4096 512 256 512 300 and GT I f : 300 512 256 512 4096).