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/

Zendy Plus

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

Zendy Tools

Zendy Open

Most models tasked to ground referential utterances in 2D and 3D scenes learnto select the referred object from a pool of object proposals provided by apre-trained detector. This is limiting because an utterance may refer to visualentities at various levels of granularity, such as the chair, the leg of thechair, or the tip of the front leg of the chair, which may be missed by thedetector. We propose a language grounding model that attends on the referentialutterance and on the object proposal pool computed from a pre-trained detectorto decode referenced objects with a detection head, without selecting them fromthe pool. In this way, it is helped by powerful pre-trained object detectorswithout being restricted by their misses. We call our model Bottom Up Top DownDEtection TRansformers (BUTD-DETR) because it uses both language guidance (topdown) and objectness guidance (bottom-up) to ground referential utterances inimages and point clouds. Moreover, BUTD-DETR casts object detection asreferential grounding and uses object labels as language prompts to be groundedin the visual scene, augmenting supervision for the referential grounding taskin this way. The proposed model sets a new state-of-the-art across popular 3Dlanguage grounding benchmarks with significant performance gains over previous3D approaches (12.6% on SR3D, 11.6% on NR3D and 6.3% on ScanRefer). Whenapplied in 2D images, it performs on par with the previous state of the art. Weablate the design choices of our model and quantify their contribution toperformance. Our code and checkpoints can be found at the project websitehttps://butd-detr.github.io.

Bottom Up Top Down Detection Transformers for Language Grounding in Images and Point Clouds