Loading...
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 53155 (05)
- University: Sharif University of Technology
- Department: Electrical Engineering
- Advisor(s): Amini, Arash; Mohammadzadeh, Nargesolhoda
- Abstract:
- High dynamic range (HDR) images provide more realistic experience in displaying real-world scenes than conventional low dynamic range (LDR) images by providing much more detailed luminance information; However, most imaging content is still available in low dynamic range. Inverse tonemapping is known as the problem of inferring an HDR image from a single-exposure LDR image in which the lost data caused by saturation of bright parts and quantization must be reconstructed.To address this problem, in this thesis, two fully-automatic architectures based on convolutional neural networks, are proposed. Both these architectures utilize a number of convolutional auto-encoders as sub-etworks.Considering that inverse tonemapping is known as an ill-posed problem, the proposed methods are indirect ones; The key idea of both methods is to produce a set of LDR images as a simulation of a real set taken with different exposures using convolutional auto-encoders as sub-networks, and then reconstruct the HDR output image by merging the resulting LDR output of each sub-network, using one of the many existing merging algorithms. One of the proposed operators constructs upper and lower exposures directly from the input LDR image; the other one utilizes a chain structure of auto-encoders. The efficacy of the proposed methods is shown by comparing the results with some conventional methods and existing CNN-based architectures
- Keywords:
- Deep Learning ; Convolutional Neural Network ; Autoencoder ; High Dynamic Range Imaging ; Tone Mapping Operators ; Low Dynamic Range (LDR)Imaging
- محتواي کتاب
- view