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 Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
HAIF-GS: Hierarchical and Induced Flow-Guided Gaussian Splatting for Dynamic Scene
Authors: Jianing Chen, Zehao Li, Yujun Cai, Hao Jiang, Chengxuan Qian, Juyuan Kang, Shuqin Gao, Honglong Zhao, Tianlu Mao, Yucheng Zhang
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on synthetic and real-world benchmarks validate that HAIF-GS significantly outperforms prior dynamic 3DGS methods in rendering quality, temporal coherence, and reconstruction efficiency. |
| Researcher Affiliation | Academia | Jianing Chen1,2, Zehao Li1,2, Yujun Cai3, Hao Jiang1,2 , Chengxuan Qian4 Juyuan Kang1,2, Shuqin Gao1, Honglong Zhao1, Tianlu Mao1,2, Yucheng Zhang1,2 1Institute of Computing Technology, Chinese Academy of Sciences, ICT 2University of Chinese Academy of Sciences, UCAS 3The University of Queensland 4Jiangsu University |
| Pseudocode | No | The paper describes methodologies using prose and mathematical equations but does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The code and models are not publicly released at the time of submission. However, we are committed to open access and will release the code, models, and reproduction instructions in the future, to enable full reproducibility of the main experimental results. |
| Open Datasets | Yes | Datasets and Metrics. We evaluate our method on two widely used benchmarks for monocular dynamic scene reconstruction: Ne RF-DS [42] and D-Ne RF [31]. |
| Dataset Splits | Yes | The Ne RF-DS dataset ... We train the model using images from the left camera and test it on the right camera. The D-Ne RF dataset ... We use the default resolution 800x800 for all scenes for training and testing. |
| Hardware Specification | Yes | All implementations are based on Py Torch framework and trained on a single NVIDIA RTX 3090 GPU. |
| Software Dependencies | No | All implementations are based on Py Torch framework and trained on a single NVIDIA RTX 3090 GPU. The paper mentions a framework but no specific version numbers for software dependencies. |
| Experiment Setup | Yes | Optimization. ... The entire training process is conducted using the Adam optimizer on a single RTX 3090 GPU. The loss weights in Equation 9 are set to ̸ = 0.8, ̸1 = 0.01, ̸2 = 0.2, and ̸3 = 0.5. |