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

Deep neural network models have achieved remarkable progress in 3D sceneunderstanding while trained in the closed-set setting and with full labels.However, the major bottleneck for current 3D recognition approaches is thatthey do not have the capacity to recognize any unseen novel classes beyond thetraining categories in diverse kinds of real-world applications. In themeantime, current state-of-the-art 3D scene understanding approaches primarilyrequire high-quality labels to train neural networks, which merely perform wellin a fully supervised manner. This work presents a generalized and simpleframework for dealing with 3D scene understanding when the labeled scenes arequite limited. To extract knowledge for novel categories from the pre-trainedvision-language models, we propose a hierarchical feature-aligned pre-trainingand knowledge distillation strategy to extract and distill meaningfulinformation from large-scale vision-language models, which helps benefit theopen-vocabulary scene understanding tasks. To leverage the boundaryinformation, we propose a novel energy-based loss with boundary awarenessbenefiting from the region-level boundary predictions. To encourage latentinstance discrimination and to guarantee efficiency, we propose theunsupervised region-level semantic contrastive learning scheme for pointclouds, using confident predictions of the neural network to discriminate theintermediate feature embeddings at multiple stages. Extensive experiments withboth indoor and outdoor scenes demonstrated the effectiveness of our approachin both data-efficient learning and open-world few-shot learning. All codes,models, and data are made publicly available at:https://drive.google.com/drive/folders/1M58V-PtR8DBEwD296zJkNg_m2qq-MTAP?usp=sharing.

Generalized Label-Efficient 3D Scene Parsing via Hierarchical Feature Aligned Pre-Training and Region-Aware Fine-tuning