InstanceFormer: An Online Video Instance Segmentation Framework
Authors: Rajat Koner, Tanveer Hannan, Suprosanna Shit, Sahand Sharifzadeh, Matthias Schubert, Thomas Seidl, Volker Tresp
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate Instance Former on three challenging datasets OVIS (Qi et al. 2021), YTVIS19 (Yang et al. 2019), and YTVIS-21 (Xu et al. 2021). ... Here, we analyze the quantitative and qualitative findings of our experiments across multiple dataset. |
| Researcher Affiliation | Academia | 1 Ludwig Maximilian University of Munich 2 Technical University of Munich 3 MCML |
| Pseudocode | No | The paper describes its methodology in narrative text and with equations, but it does not include a clearly labeled pseudocode block or algorithm. |
| Open Source Code | Yes | Code is available at https: //github.com/rajatkoner08/Instance Former. |
| Open Datasets | Yes | We evaluate Instance Former on three challenging datasets OVIS (Qi et al. 2021), YTVIS19 (Yang et al. 2019), and YTVIS-21 (Xu et al. 2021). ... We first pre-train our network on COCO (Lin et al. 2014) for 12 epochs. |
| Dataset Splits | Yes | We evaluate our model on the official validation split. |
| Hardware Specification | Yes | We train our network using the Adam W optimizer on 4 NVIDIA RTX A6000 GPUs with a batch size of 4 and learning rate of 1e 4 for 16 epochs. |
| Software Dependencies | No | The paper mentions using specific models and optimizers (e.g., 'Deformable-DETR', 'ResNet-50', 'Adam W optimizer') but does not specify software library version numbers (e.g., 'PyTorch 1.9') which would be needed for full reproducibility. |
| Experiment Setup | Yes | We train our network using the Adam W optimizer on 4 NVIDIA RTX A6000 GPUs with a batch size of 4 and learning rate of 1e 4 for 16 epochs. ... In our experiments, we have set the size of memory queue d = 4 and number of memory token per frame to 10. |