Corsmal Containers Is An Image Dataset With 1,656 Images Of 23 Transparent And Non Transparent Containers, Such As Drinking Cups, Drinking Glasses And Bottles. The Containers Have Different Shapes, Materials And Degrees Of Transparency. Each Container Stands On A Table In Two Different Setups With Different Lighting And Background Conditions, And Is Observed By Two Cameras Placed On The Top Of A Table With A Wide Baseline, Acquiring Rgb, Depth, And Narrow Baseline Stereo Infrared Images. Images Are Acquired With 1280x720 Pixels Resolution, And Rgb And Depth Images Being Spatially Aligned. The Cameras Are Calibrated And Localised With Respect To A Calibration Board. Data Acquisition Was Performed In Two Separate Rooms With Different Lighting Conditions And Different Backgrounds. Backgrounds Vary By Using Two Tablecloths In Addition To A No Tabletop Scenario. For The First Room, The Data Is Acquired Under Natural Light And The Dimensions Of The Table Top Are 160x80 Cm And The Height Of The Table Is 82 Cm From The Ground. For The Second Room (No Windows), The Data Is Acquired With Illumination Provided By Either Ceiling Lights Or Additional Portable Lights, And The Table Is Of Size 60x60 Cm And Height 82 Cm From The Ground. The Total Number Of Configurations Is 207, As The Combination Of Objects (23), Backgrounds (3) And Lighting Conditions (3), Resulting In 414 Rgb Images, 414 Depth Images And 828 Ir Images. Annotations Include The Maximum Width And Height Of Each Object. The Dataset Is Licensed Under The Creative Commons Attribution Non Commercial 4.0 International License. published 2020 via Queen Mary University of London