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

Among the most important prerequisites for creating and evaluating 6D objectpose detectors are datasets with labeled 6D poses. With the advent of deeplearning, demand for such datasets is growing continuously. Despite the factthat some of exist, they are scarce and typically have restricted setups, suchas a single object per sequence, or they focus on specific object types, suchas textureless industrial parts. Besides, two significant components are oftenignored: training using only available 3D models instead of real data andscalability, i.e. training one method to detect all objects rather thantraining one detector per object. Other challenges, such as occlusions,changing light conditions and changes in object appearance, as well preciselydefined benchmarks are either not present or are scattered among differentdatasets. In this paper we present a dataset for 6D pose estimation that coversthe above-mentioned challenges, mainly targeting training from 3D models (bothtextured and textureless), scalability, occlusions, and changes in lightconditions and object appearance. The dataset features 33 objects (17 toy, 8household and 8 industry-relevant objects) over 13 scenes of variousdifficulty. We also present a set of benchmarks to test various desireddetector properties, particularly focusing on scalability with respect to thenumber of objects and resistance to changing light conditions, occlusions andclutter. We also set a baseline for the presented benchmarks using astate-of-the-art DPOD detector. Considering the difficulty of making suchdatasets, we plan to release the code allowing other researchers to extend thisdataset or make their own datasets in the future.

HomebrewedDB: RGB-D Dataset for 6D Pose Estimation of 3D Objects