The dataset used in this contest was acquired using three XIMEA snapshot cameras including VIS, NIR and RedNIR, which respectively covers 16 bands, 25 bands and 15 bands after calibration. The videos were captured at 25 frames per second (FPS). Each frame was originally captured in 2D and arranged in a mosaic mode. Each frame is then converted to 3D with the first two dimensions index the location of each pixel, and the third dimension indexes the band number (code provided). For the RedNIR data, please drop out the last band which contains only zero values. False-color videos generated from the hyperspectral videos are also provided. The first frames of training dataset are as follows:
Dataset and Source Code Links
Dataset Link: the dataset can be accessed via Dropbox or Google Drive or Baidu YunPan Access code: 1234
Evaluation code: Evaluation.zip
2D image to hyperspectral cube conversion code: X2Cube.zip
Python code for X2Cube and Evaluation HyperTools.py
Dataset Link: the dataset can be accessed via Google Drive or Baidu YunPan Access code: 1234
The camera calibration process involves two steps: dark calibration and spectral correction. Dark calibration aims to remove the influence of noise produced by the camera sensor. It is done by subtracting a dark frame from the captured image, for which the dark frame was captured with lens covered by a cap. The goal of spectral calibration is to reduce the distortion of spectral responses. It is done by applying a sensor-specific spectral correction matrix on the measurement in each pixel. The 16 band images of the corrected hyperspectral data cube are saved as 2D frames with the 16 bands arranged again in a mosaic mode.
To ensure fair comparison, the hyperspectral videos were converted to false-color videos using CIE color matching functions. This produces strictly spatially aligned hyperspectral and false-color videos.
It is important to note that the spectral reflectance of an object may not necessarily match its actual physical reflectance due to the absence of certain calibration procedures, such as white calibration. However, despite this limitation, the spectral differences between different objects still play a significant role in improving recognition performance.
For details on the above steps, please refer to: F. Xiong, J. Zhou, and Y. Qian. "Material based object tracking in hyperspectral videos", IEEE Trans. Image Process., vol. 29, no. 1, pp. 3719-3733, 2020.
A single upright bounding box is provided for the location of the target object in each frame. The bounding box is represented by the centre location and its height and width. The labels for hyperspectral and color videos were generated independently. The labels for the hyperspectral videos can be used directly on the false-color videos.
The whole dataset contains 95 sets of videos for training and 77 sets of videos for validation. Every video is labelled with associated challenging factors out of eleven attributes, including illumination variation, scale variation, occlusion, deformation, motion blur, fast motion, in-plane rotation, out-of-plane rotation, out-of-view , background clutters, and low resolution.