Progressive Painterly Image Harmonization from Low-Level Styles to High-Level Styles

Authors: Li Niu, Yan Hong, Junyan Cao, Liqing Zhang

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on the benchmark dataset demonstrate the effectiveness of our progressive harmonization network.
Researcher Affiliation Academia Mo E Key Lab of Artificial Intelligence, Shanghai Jiao Tong University {ustcnewly, hy2628982280, joy c1, lqzhang}@sjtu.edu.cn
Pseudocode No The paper describes its network architecture and algorithms in prose and figures, but it does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement about making its source code open or available, nor does it provide a link to a code repository.
Open Datasets Yes Following previous works (Peng, Wang, and Wang 2019; Cao, Hong, and Niu 2023), we conduct experiments on COCO (Lin et al. 2014) and Wiki Art (Nichol 2016). COCO (Lin et al. 2014) contains instance segmentation annotations for 80 object categories, while Wiki Art (Nichol 2016) contains digital artistic paintings from different styles.
Dataset Splits No The paper mentions using a 'training set' and 'test set' for training the GRU and evaluating the model, and describes how composite images were created, but it does not provide specific percentages or sample counts for how the main dataset was split into training, validation, and test sets for model development.
Hardware Specification Yes Our model is implemented by Py Torch 1.10.0, which is distributed on ubuntu 20.04 LTS operation system, with 128GB memory, Intel(R) Xeon(R) Silver 4116 CPU, and one Ge Force RTX 3090 GPU.
Software Dependencies Yes Our model is implemented by Py Torch 1.10.0, which is distributed on ubuntu 20.04 LTS operation system
Experiment Setup Yes We resize the input images as 256 256 and set the batch size as 4 for model training. We adopt Adam (Kingma and Ba 2015) with learning rate of 0.0001 as the optimization solver.