DifNet: Semantic Segmentation by Diffusion Networks
Authors: Peng Jiang, Fanglin Gu, Yunhai Wang, Changhe Tu, Baoquan Chen
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The proposed Dif Net consistently produces improvements over the baseline models with the same depth and with the equivalent number of parameters, and also achieves promising performance on Pascal VOC and Pascal Context dataset. |
| Researcher Affiliation | Academia | Peng Jiang 1 Fanglin Gu 1 Yunhai Wang 1 Changhe Tu 1 Baoquan Chen 2,1 1Shandong University, China 2Peking University, China |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access to source code for the methodology described, nor does it explicitly state that the code is being released. |
| Open Datasets | Yes | We study the performance and mechanism of our Dif Net on the prevalent used Augmented Pascal VOC 2012 dataset [23, 24] and Pascal Context dataset [25]. |
| Dataset Splits | Yes | Augmented Pascal VOC 2012 dataset has 10,582 training, 1,449 validation, and 1,456 testing images with pixel-level labels in 20 foreground object classes and one background class, while Pascal Context has 4998 training and 5105 validation images with pixel-level labels in 59 classes and one background category. |
| Hardware Specification | Yes | For inference, total time consumption of our Dif Net-50 is equivalent to Sim Deeplab-101. However, in this case, the data can flow through two branches of our model parallelly, so the computation of our model can be further accelerated by model parallel to two times faster. The diffusion process only involves matrix multiplication (five random walks) and can be implemented efficiently with little extra computation. on one GTX 1080 GPU. |
| 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, PyTorch 1.9, etc.) needed to replicate the experiment. |
| Experiment Setup | Yes | To train our model and baseline models, we use a mini-batch of 16 images for 200 epochs and set learning rate, learning policy, momentum and weight decay same as [3]. We also augment training dataset by flipping, scaling and finally cropping to 321 321 due to computing resource limitation. |