DiffusionEdge: Diffusion Probabilistic Model for Crisp Edge Detection
Authors: Yunfan Ye, Kai Xu, Yuhang Huang, Renjiao Yi, Zhiping Cai
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on four edge detection benchmarks demonstrate the superiority of Diffusion Edge both in correctness and crispness. On the NYUDv2 dataset, compared to the second best, we increase the ODS, OIS (without post-processing) and AC by 30.2%, 28.1% and 65.1%, respectively. |
| Researcher Affiliation | Academia | Yunfan Ye1,2*, Kai Xu2*, Yuhang Huang2 , Renjiao Yi2, Zhiping Cai2 1Hunan University 2National University of Defense Technology |
| Pseudocode | No | The paper describes methods using equations and text, but it does not contain a clearly labeled 'Pseudocode' or 'Algorithm' block or figure. |
| Open Source Code | Yes | Code: https://github.com/Gu Huang AI/Diffusion Edge. |
| Open Datasets | Yes | We conduct experiments on four popular edge detection datasets: BSDS (Arbelaez et al. 2010), NYUDv2 (Silberman et al. 2012), Multicue (M ely et al. 2016) and BIPED (Poma, Riba, and Sappa 2020). |
| Dataset Splits | Yes | BSDS consists of 200, 100, and 200 images in the training set, validation set, and test set, respectively. ... NYUDv2 ... is divided into 381 training, 414 validation and 654 testing images. ... Multicue ... We randomly split the 100 images into training and evaluation sets, consisting of 80 and 20 images respectively. ... BIPED contains 250 annotated images of outdoor scenes and is split into a training set of 200 images and a testing set of 50 images. |
| Hardware Specification | Yes | All the training is conducted on a single RTX 3090 GPU. When inferencing each single image on BSDS dataset, with the sampling Equation 2, it takes about 3.5GB GPU memory, 1.2 seconds for one-step sampling and 3.2 seconds for five steps on a 3080Ti GPU. |
| Software Dependencies | No | The paper states 'We implement our Diffusion Edge using Py Torch (Paszke et al. 2019)', but does not provide specific version numbers for PyTorch or any other software dependencies. |
| Experiment Setup | Yes | We train the models using Adam W optimizer with an attenuated learning rate (from 5e 5 to 5e 6) for 25k iterations, and each training takes up about 15 GPU hours. We employ the exponential moving average (EMA) to prevent unstable model performances during the training process. The balancing weight λ and the threshold η to identify uncertain edge pixels are set to 1.1 and 0.3, respectively, for all experiments. We train all datasets with randomly cropped patches of size 320 320 with batch size 16. |