Image Composition with Depth Registration

Authors: Zan Li, Wencheng Wang, Fei Hou

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that we can conveniently handle occlusions in image composition and improve efficiency by about 4 times compared to Photoshop. Experimental results were collected on a personal computer installed with an Intel(R) Core(TM) i7-6700 CPU, 16GB RAM, and Windows 10.
Researcher Affiliation Academia 1State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences 2University of Chinese Academy of Sciences, Beijing, China
Pseudocode No The algorithm steps are described in Section 4 ('Algorithm for Image Composition'), but they are presented as a numbered list within the regular text rather than a formal pseudocode block or algorithm environment.
Open Source Code No The paper does not provide a link to the open-source code for its described methodology, nor does it explicitly state that the code is publicly available.
Open Datasets No The paper states, 'the Photoshop compositions are used as the ground truth as they are manually produced,' indicating the use of a dataset. However, it does not provide any specific access information (e.g., URL, DOI, specific repository, or citation to a public dataset) for this data.
Dataset Splits No The paper does not explicitly provide details about training, validation, or test dataset splits (e.g., percentages, sample counts, or citations to predefined splits) that would be needed for reproduction.
Hardware Specification Yes Experimental results were collected on a personal computer installed with an Intel(R) Core(TM) i7-6700 CPU, 16GB RAM, and Windows 10.
Software Dependencies Yes The proposed method was implemented in Python 3.8.9. For comparison, we implemented the method in [Tan et al., 2019] in Visual C++ and downloaded the learned network (Comp GAN) of [Azadi et al., 2020]. Semantic segmentation uses primarily the popular method Deep Lab V3+ [Chen et al., 2018]. We also use Grab Cut [Rother et al., 2004] to further optimize the extracted objects. Depth estimation uses Ne WCRFs [Yuan et al., 2022] to obtain a high-resolution depth map given a single RGB image. Color blending uses the enhanced matting method of [Wang et al., 2016].
Experiment Setup No The paper describes the general implementation details, such as the use of Deep Lab V3+ for semantic segmentation and NeWCRFs for depth estimation, along with some details about color blending (e.g., 'weight wi is set to 0.8 for the pixels on the boundary...'), but it does not provide specific hyperparameters for model training (e.g., learning rate, batch size, number of epochs) or other system-level training settings for its overall composition method.