Recovering high dynamic range radiance maps from photographs
Proceedings of the 24th annual conference on Computer graphics and interactive techniques
Determining the Camera Response from Images: What Is Knowable?
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic High-Dynamic Range Image Generation for Dynamic Scenes
IEEE Computer Graphics and Applications
| Hi-index | 0.03 |
Ghosting artifacts could be generated in a high dynamic range (HDR) image if it is synthesized for a scene with moving objects by using differently exposed low dynamic range (LDR) images. To remove the ghosting artifacts, pixels affected by movement, i.e., moving regions, are detected from all LDR images, and all pixels in the moving regions are not involved in the synthesis of the final HDR image. Besides removing the ghosting artifacts. it is also very important to preserve the dynamic ranges of moving objects. A fast intensity mapping function (IMF) based method is introduced in this poster to patch the moving regions. With the proposed method, all LDR images are involved in the patching of moving regions instead of only one LDR image as in existing methods [Jacobs et al. 2008]. Therefore, the dynamic range of the HDR image is preserved better than existing methods.