Deep Optical Flow Estimation Via Multi-Scale Correspondence Structure Learning

Authors: Shanshan Zhao, Xi Li, Omar El Farouk Bourahla

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on several benchmark datasets demonstrate the effectiveness of the proposed approach.
Researcher Affiliation Collaboration 1 Zhejiang University, Hangzhou, China 2 Alibaba-Zhejiang University Joint Institute of Frontier Technologies, Hangzhou, China
Pseudocode Yes Algorithm 1: Deep Optical Flow Estimation Via MSCSL
Open Source Code No The paper does not provide any statement or link regarding the availability of its source code.
Open Datasets Yes Flying Chairs [Dosovitskiy et al., 2015] is a synthetic dataset... MPI Sintel [Butler et al., 2012] is created from an animated movie... KITTI 2012 [Geiger et al., 2012] is created from real world scenes... Middlebury [Baker et al., 2011] is a very small dataset...
Dataset Splits Yes Flying Chairs [...] is split into 22, 232 training and 640 test pairs. [...] we fine-tune the networks on the Clean version and Final version together with 1, 816 for training and 266 for validation.
Hardware Specification Yes We implement our architecture using Caffe [Jia et al., 2014] and use an NVIDIA TITAN X GPU to train the network.
Software Dependencies No The paper mentions
Experiment Setup Yes We train the networks on Flying Chairs training dataset using Adam optimization with β1 = 0.9 and β2 = 0.999. To tackle the gradients explosion, we adopt the same strategy as proposed in [Dosovitskiy et al., 2015]. Specifically, we firstly use a learning rate of 1e 6 for the first 10k iterations with a batch size of 8 pairs. After that, we increase the learning rate to 1e 4 for the following 300k iterations, and then divide it by 2 every 100k iterations. We terminate the training after 600k iterations (about 116 hours).