SOccDPT: Semi-Supervised 3D Semantic Occupancy from Dense Prediction Transformers trained under memory constraints
This addresses 3D scene understanding for autonomous vehicles in unstructured traffic environments, though it appears incremental with specific optimizations.
The authors tackled 3D semantic occupancy prediction from monocular images in unstructured traffic scenarios by developing SOccDPT, a memory-efficient dense prediction transformer trained semi-supervised on Indian datasets, achieving an RMSE of 9.1473, IoU of 46.02%, and 69.47 Hz inference.
We present SOccDPT, a memory-efficient approach for 3D semantic occupancy prediction from monocular image input using dense prediction transformers. To address the limitations of existing methods trained on structured traffic datasets, we train our model on unstructured datasets including the Indian Driving Dataset and Bengaluru Driving Dataset. Our semi-supervised training pipeline allows SOccDPT to learn from datasets with limited labels by reducing the requirement for manual labelling by substituting it with pseudo-ground truth labels to produce our Bengaluru Semantic Occupancy Dataset. This broader training enhances our model's ability to handle unstructured traffic scenarios effectively. To overcome memory limitations during training, we introduce patch-wise training where we select a subset of parameters to train each epoch, reducing memory usage during auto-grad graph construction. In the context of unstructured traffic and memory-constrained training and inference, SOccDPT outperforms existing disparity estimation approaches as shown by the RMSE score of 9.1473, achieves a semantic segmentation IoU score of 46.02% and operates at a competitive frequency of 69.47 Hz. We make our code and semantic occupancy dataset public.