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 TransProteus, a dataset, and methods for predicting the 3Dstructure, masks, and properties of materials, liquids, and objects insidetransparent vessels from a single image without prior knowledge of the imagesource and camera parameters. Manipulating materials in transparent containersis essential in many fields and depends heavily on vision. This work supplies anew procedurally generated dataset consisting of 50k images of liquids andsolid objects inside transparent containers. The image annotations include 3Dmodels, material properties (color/transparency/roughness...), and segmentationmasks for the vessel and its content. The synthetic (CGI) part of the datasetwas procedurally generated using 13k different objects, 500 differentenvironments (HDRI), and 1450 material textures (PBR) combined with simulatedliquids and procedurally generated vessels. In addition, we supply 104real-world images of objects inside transparent vessels with depth maps of boththe vessel and its content. We propose a camera agnostic method that predicts3D models from an image as an XYZ map. This allows the trained net to predictthe 3D model as a map with XYZ coordinates per pixel without prior knowledge ofthe image source. To calculate the training loss, we use the distance betweenpairs of points inside the 3D model instead of the absolute XYZ coordinates.This makes the loss function translation invariant. We use this to predict 3Dmodels of vessels and their content from a single image. Finally, wedemonstrate a net that uses a single image to predict the material propertiesof the vessel content and surface.

Predicting 3D shapes, masks, and properties of materials, liquids, and objects inside transparent containers, using the TransProteus CGI dataset