Semantic-Aware Data Augmentation for Text-to-Image Synthesis
Authors: Zhaorui Tan, Xi Yang, Kaizhu Huang
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments validate that SADA enhances text-image consistency and improves image quality significantly in T2Isyn models across various backbones. |
| Researcher Affiliation | Academia | Zhaorui Tan1,2, Xi Yang1 , Kaizhu Huang3* 1Department of Intelligent Science, Xi an Jiaotong-Liverpool University 2Department of Computer Science, University of Liverpool 3 Data Science Research Center, Duke Kunshan University |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks clearly labeled as "Pseudocode" or "Algorithm". |
| Open Source Code | Yes | Codes are available at https: //github.com/zhaorui-tan/SADA. |
| Open Datasets | Yes | Datasets CUB (Wah et al. 2011), COCO (Lin et al. 2014), MNIST, and Pok emon BLIP (Deng 2012) are employed for training and tuning |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning in the main text, only mentioning "training and tuning". |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | No | The paper states "Parameter settings follow the original models of each framework for all experiments unless specified" but does not provide specific experimental setup details (concrete hyperparameter values, training configurations, or system-level settings) in the main text. |