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

3D vision-language grounding (3D-VL) is an emerging field that aims toconnect the 3D physical world with natural language, which is crucial forachieving embodied intelligence. Current 3D-VL models rely heavily onsophisticated modules, auxiliary losses, and optimization tricks, which callsfor a simple and unified model. In this paper, we propose 3D-VisTA, apre-trained Transformer for 3D Vision and Text Alignment that can be easilyadapted to various downstream tasks. 3D-VisTA simply utilizes self-attentionlayers for both single-modal modeling and multi-modal fusion without anysophisticated task-specific design. To further enhance its performance on 3D-VLtasks, we construct ScanScribe, the first large-scale 3D scene-text pairsdataset for 3D-VL pre-training. ScanScribe contains 2,995 RGB-D scans for 1,185unique indoor scenes originating from ScanNet and 3R-Scan datasets, along withpaired 278K scene descriptions generated from existing 3D-VL tasks, templates,and GPT-3. 3D-VisTA is pre-trained on ScanScribe via masked language/objectmodeling and scene-text matching. It achieves state-of-the-art results onvarious 3D-VL tasks, ranging from visual grounding and dense captioning toquestion answering and situated reasoning. Moreover, 3D-VisTA demonstratessuperior data efficiency, obtaining strong performance even with limitedannotations during downstream task fine-tuning.

3D-VisTA: Pre-trained Transformer for 3D Vision and Text Alignment