Self-Supervised Object Localization with Joint Graph Partition
Authors: Yukun Su, Guosheng Lin, Yun Hao, Yiwen Cao, Wenjun Wang, Qingyao Wu2289-2297
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To validate the effectiveness of the proposed method, extensive experiments are conducted on CUB-200-2011, Stanford Cars and FGVC-Aircraft datasets. Experimental results show that our method outperforms state-of-the-art methods using the same level of supervision, even outperforms some weakly-supervised methods. |
| Researcher Affiliation | Academia | Yukun Su1,2, Guosheng Lin2 , Yun Hao1,3, Yiwen Cao1,3, Wenjun Wang1,3, Qingyao Wu1,4 1School of Software and Engineering, South China University of Technology 2School of Computer Science and Engineering, Nanyang Technological University 3Key Laboratory of Big Data and Intelligent Robot, Ministry of Education 4 Pazhou Lab, Guangzhou, China |
| Pseudocode | No | The paper describes the methodology in prose and through diagrams (Figure 3, Figure 4) but does not include any explicit pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain an explicit statement about releasing source code or a link to a code repository for the described methodology. |
| Open Datasets | Yes | To evaluate the proposed approach, three datasets are adopted, including CUB-200-2011 (Wah et al. 2011), FGVC-Aircraft (Maji et al. 2013) and Stanford Cars (Krause et al. 2013). |
| Dataset Splits | No | We strictly follow the train-list and test-list of the datasets for training and evaluation and the bounding box annotations are solely used for evaluation. Among them, CUB-200-2011 is the largest dataset that contains 200 categories of birds with 5,994 training images and 5,794 testing images. |
| Hardware Specification | Yes | In this work, we implement the proposed framework with Py Torch and train on 2080Ti-GPUs. |
| Software Dependencies | No | In this work, we implement the proposed framework with Py Torch and train on 2080Ti-GPUs. |
| Experiment Setup | Yes | The input image was resized to 256 256 and then was randomly cropped to 224 224 using zero padding if needed. We use stochastic gradient descent (SGD) optimizer with initial learning rate of 0.001, momentum of 0.9 and batch size of 32 for the model. The weight decay is set to 0.004. |