English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present SOccDPT, a memory-efficient approach for 3D semantic occupancyprediction from monocular image input using dense prediction transformers. Toaddress the limitations of existing methods trained on structured trafficdatasets, we train our model on unstructured datasets including the IndianDriving Dataset and Bengaluru Driving Dataset. Our semi-supervised trainingpipeline allows SOccDPT to learn from datasets with limited labels by reducingthe requirement for manual labelling by substituting it with pseudo-groundtruth labels to produce our Bengaluru Semantic Occupancy Dataset. This broadertraining enhances our model's ability to handle unstructured traffic scenarioseffectively. To overcome memory limitations during training, we introducepatch-wise training where we select a subset of parameters to train each epoch,reducing memory usage during auto-grad graph construction. In the context ofunstructured traffic and memory-constrained training and inference, SOccDPToutperforms existing disparity estimation approaches as shown by the RMSE scoreof 9.1473, achieves a semantic segmentation IoU score of 46.02% and operates ata competitive frequency of 69.47 Hz. We make our code and semantic occupancydataset public.

SOccDPT: Semi-Supervised 3D Semantic Occupancy from Dense Prediction Transformers trained under memory constraints