Motion-Attentive Transition for Zero-Shot Video Object Segmentation
Authors: Tianfei Zhou, Shunzhou Wang, Yi Zhou, Yazhou Yao, Jianwu Li, Ling Shao13066-13073
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on three challenging public benchmarks (i.e. DAVIS-16, FBMS and Youtube-Objects) show that our model achieves compelling performance against the state-of-the-arts. |
| Researcher Affiliation | Collaboration | 1Inception Institute of Artifical Intelligence, UAE 2Beijing Key Laboratory of Intelligent Information Technology, School of Computer Science and Technology, Beijing Institute of Technology, China 3School of Computer Science and Engineering, Nanjing University of Science and Technology, China |
| Pseudocode | No | The paper describes algorithms and models using equations and diagrams, but does not provide structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code is available at: https://github.com/tfzhou/MATNet. |
| Open Datasets | Yes | Our training data consist of two parts: i) all training data in DAVIS-16 (Perazzi et al. 2016)... ii) a subset of 12K frames selected from the training set of Youtube-VOS (Xu et al. 2018) |
| Dataset Splits | Yes | DAVIS-16 consists of 50 high-quality video sequences (30 for training and 20 for validation) in total. |
| Hardware Specification | Yes | all the experiments are conducted using a single Nvidia RTX 2080Ti GPU and an Intel(R) Xeon Gold 5120 CPU. |
| Software Dependencies | No | The paper mentions 'implemented with Py Torch' but does not provide a version number for PyTorch or any other software dependencies. |
| Experiment Setup | Yes | The entire network is trained using the SGD optimizer with an initial learning rate of 1e-4 for the encoder and the bridge network, and 1e-3 for the decoder. During training, the batch size, momentum and weight decay are set to 2, 0.9, and 1e5, respectively. The data are augmented online with horizontal flip and rotations covering a range of degrees ( 10, 10). |