Login
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision | |||
|
research:monorec [2021/06/15 09:32] Nan Yang |
research:monorec [2021/06/15 09:38] (current) Nan Yang |
||
|---|---|---|---|
| Line 2: | Line 2: | ||
| **Contact: | **Contact: | ||
| + | |||
| + | < | ||
| + | [June 14., 2021] Code released! See below. | ||
| + | </ | ||
| < | < | ||
| Line 18: | Line 22: | ||
| In this paper, we propose MonoRec, a semi-supervised monocular dense reconstruction architecture that predicts depth maps from a single moving camera in dynamic environments. MonoRec is based on a multi-view stereo setting which encodes the information of multiple consecutive images in a cost volume. To deal with dynamic objects in the scene, we introduce a MaskModule that predicts moving object masks by leveraging the photometric inconsistencies encoded in the cost volumes. Unlike other multi-view stereo methods, MonoRec is able to predict accurate depths for both static and moving objects by leveraging the predicted masks. Furthermore, | In this paper, we propose MonoRec, a semi-supervised monocular dense reconstruction architecture that predicts depth maps from a single moving camera in dynamic environments. MonoRec is based on a multi-view stereo setting which encodes the information of multiple consecutive images in a cost volume. To deal with dynamic objects in the scene, we introduce a MaskModule that predicts moving object masks by leveraging the photometric inconsistencies encoded in the cost volumes. Unlike other multi-view stereo methods, MonoRec is able to predict accurate depths for both static and moving objects by leveraging the predicted masks. Furthermore, | ||
| + | |||
| + | ==== Code ==== | ||
| + | The source code of MonoRec has been released on GitHub under the flexible MIT License: https:// | ||
| ==== Publications ==== | ==== Publications ==== | ||
