Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
BLiRF: Bandlimited Radiance Fields for Dynamic Scene Modeling
Authors: Sameera Ramasinghe, Violetta Shevchenko, Gil Avraham, Anton van den Hengel
AAAI 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically validate the efficacy of our framework by demonstrating better modeling of long-range dynamics, motion localization, and light/texture changes, achieving state-of-the-art results over competitive datasets. Our model only takes around 3 hours per scene to train (more than 10 times faster than NR-Ne RF, D-Ne RF, and Hyper Ne RF), and does not require complex loss regularizers or optimization procedures. |
| Researcher Affiliation | Industry | Sameera Ramasinghe, Violetta Shevchenko, Gil Avraham, Anton van den Hengel Amazon, Australia |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating that source code for the methodology is available. |
| Open Datasets | Yes | To demonstrate the ability of our method to capture topologically varying deformations, we also evaluate against the Hyper Ne RF dataset (Park et al. 2021b). |
| Dataset Splits | No | The paper discusses training and testing but does not explicitly provide specific train/validation/test dataset splits with percentages or counts. |
| Hardware Specification | Yes | Our model only takes around 3 hours per scene to train (more than 10 times faster than NR-Ne RF, D-Ne RF, and Hyper Ne RF), and does not require complex loss regularizers or optimization procedures. ... (3 hours on a single V100) |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | No | our model uses a single hyperparameter setting across all the scenes, demonstrating its robustness (see supplementary for hyperparameter and training details). Further, it is essential to validate whether the superior performance of our model stems from the light/density disentanglement or the space-time factorization. |