Name
Title | ||
|---|---|---|
Temporal Interpolation as an Unsupervised Pretraining Task for Optical Flow Estimation. |
Abstract | ||
|---|---|---|
The difficulty of annotating training data is a major obstacle to using CNNs for low-level tasks in video. Synthetic data often does not generalize to real videos, while unsupervised methods require heuristic losses. Proxy tasks can overcome these issues, and start by training a network for a task for which annotation is easier or which can be trained unsupervised. The trained network is then fine-tuned for the original task using small amounts of ground truth data. Here, we investigate frame interpolation as a proxy task for optical flow. Using real movies, we train a CNN unsupervised for temporal interpolation. Such a network implicitly estimates motion, but cannot handle untextured regions. By fine-tuning on small amounts of ground truth flow, the network can learn to fill in homogeneous regions and compute full optical flow fields. Using this unsupervised pre-training, our network outperforms similar architectures that were trained supervised using synthetic optical flow. |
Year | Venue | DocType |
|---|---|---|
2018 | GCPR | Conference |
Volume | Citations | PageRank |
abs/1809.08317 | 1 | 0.35 |
References | Authors | |
16 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jonas Wulff | 1 | 438 | 17.59 |
| Michael J. Black | 2 | 11233 | 1536.41 |