Reliable and Calibrated Semantic Occupancy Prediction by Hybrid Uncertainty Learning
Authors: Song Wang, Zhongdao Wang, Jiawei Yu, Wentong Li, Bailan Feng, Junbo Chen, Jianke Zhu
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we conduct a comprehensive evaluation of existing semantic occupancy prediction models from a reliability perspective for the first time. ... Extensive experiments under various settings demonstrate that RELIOCC significantly enhances the reliability of learned model while maintaining the accuracy for both geometric and semantic predictions. ... Extensive experiments on online uncertainty learning and offline model calibration across diverse settings demonstrate the effectiveness of our approach in bridging the reliability gap between camera and Li DAR-based methods under general conditions, while showcasing robust performance in adverse scenarios such as sensor failures and noisy observations. |
| Researcher Affiliation | Collaboration | 1Zhejiang University 2Huawei Noah s Ark Lab 3Nanjing University of Aeronautics and Astronautics 4Udeer.ai EMAIL, EMAIL |
| Pseudocode | No | The paper describes methods and processes using textual descriptions and diagrams (e.g., Figure 1: Overview of proposed RELIOCC), but it does not contain any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain an explicit statement about releasing the source code for the described methodology, nor does it provide a link to a code repository. It mentions 'RELIOC provides a plug-and-play scheme for existing models,' but this refers to the design, not the availability of its own implementation. |
| Open Datasets | Yes | Semantic KITTI [Behley et al., 2019] is the first large-scale outdoor dataset for semantic occupancy prediction containing 64-beam Li DAR scans and front camera images as inputs [Geiger et al., 2012]. |
| Dataset Splits | Yes | The dataset comprises 22 sequences, where 00-10 (excluding 08) are used as the training set, 08 is the validation set, and 11-21 are the test set. |
| Hardware Specification | No | For both frameworks, these methods are trained on 8 GPUs for 20 epochs with a learning rate as 0.001 and Adam W optimizer [Zhao et al., 2022]. The batch size is set to 1 per GPU. While GPUs are mentioned, specific models (e.g., NVIDIA A100) are not provided. |
| Software Dependencies | No | For both frameworks, these methods are trained on 8 GPUs for 20 epochs with a learning rate as 0.001 and Adam W optimizer [Zhao et al., 2022]. The batch size is set to 1 per GPU. The paper mentions the 'Adam W optimizer' but does not specify versions for any programming languages, libraries, or frameworks like PyTorch or TensorFlow. |
| Experiment Setup | Yes | The loss coefficients (α, β) for Lau and Lru are set to (4.0, 6.0) for both frameworks. ... For both frameworks, these methods are trained on 8 GPUs for 20 epochs with a learning rate as 0.001 and Adam W optimizer [Zhao et al., 2022]. The batch size is set to 1 per GPU. ... For our method, the loss weights (α, β, γ) for uncertainty learning (Lau, Lru) and model calibration (Lcalib) are set to 1.5, 1.0, and 4.0, respectively. |