<p id="9z1vv"></p>

          <ruby id="9z1vv"></ruby>
            <ruby id="9z1vv"></ruby>

              <p id="9z1vv"></p>
              <ruby id="9z1vv"></ruby>
              <ruby id="9z1vv"></ruby>

                  Joint Reflection Removal and Depth Estimation from a Single Image

                  Yakun Chang, Cheolkon Jung, and Jun Sun

                  Xidian University

                  Left: Image with reflection. Middle: Reflection removal by the proposed method. Right: Depth estimation by the proposed method.

                  Datasets

                  (1) Real image dataset: Download(118M)

                  (2) Synthetic image dataset: Download(2.9G)

                  Results:

                  (1) Results on real image dataset:

                  Input real images with reflections

                  Transmission recovery by the proposed method

                  Ground truth color images

                  Depth estimation by the proposed method

                  Ground truth depth maps

                  (2) Results on synthetic image dataset:

                  Input synthetic images with reflections

                  Transmission recovery by the proposed method

                  Ground truth color images

                  Depth estimation by the proposed method

                  Ground truth depth maps

                  Acknowledgement

                  This work was supported by the National Natural Science Foundation of China (No. 61271298) and the International S&T Cooperation Program of China (No. 2014DFG12780)

                  ?2015 Xidian Media Lab. All rights reserved.
                  Email: jushutaoxidian@163.com

                    <p id="9z1vv"></p>

                          <ruby id="9z1vv"></ruby>
                            <ruby id="9z1vv"></ruby>

                              <p id="9z1vv"></p>
                              <ruby id="9z1vv"></ruby>
                              <ruby id="9z1vv"></ruby>

                                  伊人久久综合