INDEPENDENT MOVING OBJECTS FOR VELOCITY ESTIMATION
We introduce I-MOVE, the first publicly available RGB-D/stereo dataset for estimating velocities of independently moving objects. Velocity estimation given RGB-D data is an unsolved problem. The I-MOVE dataset provides an opportunity for generalizable velocity estimation models to be created and have their performance be accurately and fairly measured. The dataset features various outdoor and indoor scenes of single and multiple moving objects. Compared to other datasets, I-MOVE is unique because the RGB-D data and speed for each object are supplied for a variety of different settings/environments, objects, and motions. The dataset includes training and test sequences captured from four different depth camera views and three 4K-stereo setups. The data are also time-synchronized with three Doppler radars to provide the magnitude of velocity ground truth. The I-MOVE dataset includes complex scenes from moving pedestrians via walking and biking to multiple rolling objects, all captured with the seven cameras, providing over 500 Depth/Stereo videos.
How To Use
STEP 1: CHOOSE YOUR DATASET
I-MOVE captured each scene with seven different stereo set-ups. Each device varies in either a framerate, depth estimation method, or field of view. Choose the dataset you wish to work with by selecting the camera (Intel RealSense 415, Intel RealSense 435, MYNT Eye, ZED, Stereo GoPro 122, Stereo GoPro 54, Stereo GoPro 65), depth estimation method (Stereo RGB, Stereo RGB-D (IR), Monochromatic IR) or download all.
STEP 2: CREATE MODEL / TRAIN ON DATA
The dataset contains all necessary stereo data, along with intrinsic and extrinsic calibration scenes so you may alter the depth estimation method if you do not wish to use the standard calibration for the Intel RealSense and ZED devices. The dataset provides radar ground truth data and scene information for physics calculations (e.g. height of release, or angle of release) so you may use either ground truths or both for the training of your model. In order to properly evaluate your model/method, your predictions of magnitude of velocity must be in meters per second.
STEP 3: TEST/RUN EVALUATION SCRIPTS
When you download the dataset you will be provided with an evaluation directory. Run the evaluation script for each of your respective test scenes and your magnitude of velocity estimations (in m/s) will be evaluated against both the physics calculation based and radar ground truth data. These evaluations will be done using the mean absolute deviation method.
The dataset provided on this page is published under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License. This means that you must attribute the work in the manner specified by the authors, you may not use this work for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license. If you are interested in commercial usage you can contact us for further options.