A Diffusion-Based Framework for Occluded Object Movement
Authors: Zheng-Peng Duan, Jiawei Zhang, Siyu Liu, Zheng Lin, Chun-Le Guo, Dongqing Zou, Jimmy Ren, Chongyi Li
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive evaluations demonstrate the superior performance of our method, which is further validated by a comprehensive user study. |
| Researcher Affiliation | Collaboration | 1VCIP, CS, Nankai University 2Sense Time Research 3NKIARI, Shenzhen Futian 4PBVR 5BNRist, Department of Computer Science and Technology, Tsinghua University |
| Pseudocode | No | The paper describes the methodology in prose and mathematical formulations, but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper provides a project link (https://adam-duan.github.io/projects/diffoom/) but does not explicitly state that the source code for the methodology is released or provide a direct link to a code repository. |
| Open Datasets | Yes | To compare the performance of our method with other existing arts, we develop a specialized evaluation dataset derived from COCOA (Zhu et al. 2017) training and validation sets. ... For each sample, the visible mask Mv is provided by the COCO dataset (Lin et al. 2014). |
| Dataset Splits | No | The final evaluation dataset comprises 120 images with a total of 150 sample objects. For each sample, we randomly set 8 different target positions, which results in 1200 testing cases in total. However, the paper does not specify the training/validation dataset splits used for their model. |
| Hardware Specification | No | The computational devices of this work is supported by the Supercomputing Center of Nankai University (NKSC). This statement indicates the computing environment but does not provide specific hardware details such as GPU/CPU models. |
| Software Dependencies | No | The paper mentions using 'Stable Diffusion V1.5' but does not specify any programming languages, libraries, or other software dependencies with version numbers. |
| Experiment Setup | No | The paper describes various strategies and components of its methodology, such as 'Latent Hold strategy', 'Color Fill strategy', and 'Local Text Guidance', and mentions variables like 'relax ratio η', 'Tm-th step', 'guidance scale ω', and 'gradient descent parameter γ'. However, it does not provide specific numerical values for hyperparameters or system-level training settings like learning rate, batch size, or number of epochs. |