A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation

From Computer Vision Freiburg:  Recent work has shown that optical flow estimation can be formulated as a supervised learning task and can be successfully solved with convolutional networks. Training of the so-called FlowNet was enabled by a large synthetically generated dataset. The present paper extends the concept of optical flow estimation via convolutional networks to disparity and scene flow estimation. To this end, we propose three synthetic stereo video datasets with sufficient realism, variation, and size to successfully train large networks. Our datasets are the first large-scale datasets to enable training and evaluating scene flow methods. Besides the datasets, we present a convolutional network for real-time disparity estimation that provides state-of-the-art results. By combining a flow and disparity estimation network and training it jointly, we demonstrate the first scene flow estimation with a convolutional network.

his video shows impressions from various parts of our dataset, as well as state-of-the-art realtime disparity estimation results produced by one of our new CNNs... (full paper)

 

 

Comments (0)

This post does not have any comments. Be the first to leave a comment below.


Post A Comment

You must be logged in before you can post a comment. Login now.

Featured Product

maxon motor’s - Compact power for your motion control application.

maxon motor's - Compact power for your motion control application.

maxon launches the next generation of positioning controllers - the EPOS4. A high performance module with detachable pin headers and two different power ratings. With a connector board, the modules can be combined into a ready-to-install compact solution. Suitable for efficient and dynamic control of brushed and brushless DC motors with Hall sensors and encoders up to 750 W continuous power and 1500 W peak power. The modular concept also provides for a wide variety of expansion options with Ethernet-based interfaces, such as EtherCAT or absolute rotary encoders.