Learning Spatio-temporal Distortion Models for HDR Video Denoising and Deblurring
1. Additional Results for HDR Reconstruction Using Single Interleaved Exposure
This section shows additional HDR results captured with various exposure times.
BrochureCup CarCardsCeilingCandle
2. Temporal super-resolution
This section illustrates temporally created 4 sharp HDR frames using our temporal super-resolution network.
Example Secene 1:
Frame 1Frame 2Frame 3Frame 4
Example Secene 2:
Frame 1Frame 2Frame 3Frame 4
3. Blur Adjustment
This section explains continous adjustment of motion blur for the single HDR frame.
There is a continous blur information in high exposure that we cut it into individual sharp frames which emulate certain window of time.
Then, blur is achieved by averaging temporally reconstructed those sharp HDR frames. The amount of blur depends on the number of averaged frames
which means addition of all frames simulate the exact blur information in high exposure.
As a result, objects are blurred naturally along the motion direction in our reconstructions.
CupPenMarbleCandle
Blur setting:
Low/High Exposure Time: 8ms/32ms
4. Video
This section demonstrates reconstructed HDR videos captured in dynamic scenes using a synthetic data.
Video 1
5. Video
This section demonstrates reconstructed HDR videos captured in dynamic scenes. Videos are reconstructed with our temporal method using captured images.
Video 1Video 2Video 3Video 4
Low/High Exposure Time: 0.5ms/2ms/8ms
6. Input Images and Results Figure 1
Dual Sensor Capturing
Low ExposureMedium ExposureHigh ExposureKalantari et al.Yan et al.Liu et al.Ours
Conventional Sensor Capturing
Low ExposureMedium ExposureHigh ExposureKalantari et al.Yan et al.Liu et al.
7. Input Images and Results Figure 9
Dual Sensor Capturing
Low ExposureMedium ExposureHigh ExposureKalantari et al.Yan et al.Liu et al.Ours
Conventional Sensor Capturing
Low ExposureMedium ExposureHigh ExposureKalantari et al.Yan et al.Liu et al.