Home » CURE-OR-1M

CURE-OR-1M

CURE-OR: Challenging Unreal and Real Environments for Object Recognition

coca_cola_cNums_color.gif

Description

We introduced an image dataset denoted as Challenging Unreal and Real Environments for Object Recognition (CURE-OR-1M), whose characteristics are summarized below. In CURE-OR-1M dataset, there are 1,000,000 images of 100 objects with varying size, color, and texture, captured with multiple devices in different setups. In contrast to existing studies, the majority of images in the CURE-OR-1M dataset were acquired with smartphones and tested with off-the-shelf applications, because we want to benchmark the recognition performance of devices and applications that are used in our daily lives rather than testing algorithms that can only work with specific hardware or testing devices that we rarely utilize.

Dataset Characteristics

 Object Classes (number of objects/class) Number of images per object Controlled condition (level) Background Acquisition
toy (23) personal (10) office (14) household (27) sports/ent. (10) health (16) 10,000 Perspective (5) background (5) challenge (78) white 2D      textured 2D (2)  real 3D (2) DSLR     webcam smartphones

Download

Will be available soon, please stay tuned!

Publications

Coming soon…

Objects

objects

Backgrounds

backgrounds

Challenging Conditions

ctypes_clevels_nolabels2.jpg

 

%d bloggers like this: