ARMBench: An Object-centric Benchmark Dataset for Robotic Manipulation

6D-ViCuT: Six degree-of-freedom visual cuboid tracking dataset for manual packing of cargo in warehouses

“…These features are challenging to develop the 6D visual tracking systems for objects in packing applications. The evolution of trackers in the logistics field has been founded in datasets such as BigBird [1] , APC [2] , ARMBench [3] , and recent Amazon dataset used in [4] . These datasets explore a standard configuration where the cargo resides in a small container [ 1 , 2 , 3 ] or conveyor [4] reachable by an industrial manipulator.…”

“…The evolution of trackers in the logistics field has been founded in datasets such as BigBird [1] , APC [2] , ARMBench [3] , and recent Amazon dataset used in [4] . These datasets explore a standard configuration where the cargo resides in a small container [ 1 , 2 , 3 ] or conveyor [4] reachable by an industrial manipulator. None of these datasets explores the configuration of a mobile operator, simultaneously solving the exploration of a consolidation area in search of the cargo, the manual picking of the cargo, and the manual accommodation of cargo in the container [5] .…”

Dimensioning Cuboid and Cylindrical Objects Using Only Noisy and Partially Observed Time-of-Flight Data

“…Each of these ground surfaces has different levels of infrared reflectivity; (3) quantifying the accuracy for dimensioning cuboid and cylindrical objects with different rotation angles and orientations with respect to the ToF sensor; and (4) quantifying the accuracy for dimensioning cuboid and cylindrical objects using various techniques for limiting overgrowth when fitting superquadric models. In applications such as logistics, the ability to accurately dimension objects is critical for operations like object grasping, packaging, storing, and transportation [33][34][35][36][37]. The tolerances for dimensioning errors vary from system to system.…”

“…The tolerances for dimensioning errors vary from system to system. As these systems are further developed, their tolerances are typically reduced to optimize efficiency, object handling, and packaging [33][34][35][36][37]. As such, it becomes increasingly important to understand and quantify the performance and accuracy of object dimensioning techniques and the various factors that affect their performance.…”