Awesome

eSL-Net

This is code for the paper Event Enhanced High-Quality Image Recovery by Bishan Wang, Jingwei He, Lei Yu, Gui-Song Xia, Wen Yang.

You can find a pdf of the paper here. The paper has been accepted by ECCV2020. If you use of this code or the synthetic dataset, please cite the following publications:

@inproceedings{wang2020event,
  title={Event Enhanced High-Quality Image Recovery},
  author={Wang, Bishan and He, Jingwei and Yu, Lei and Xia, Gui-Song and Yang, Wen},
  booktitle={European Conference on Computer Vision},
  year={2020},
  organization={Springer}
}

Run

• Pretrained model with SR : code/pre_trained/model_withsr_pretrained.pt

• Pretrained model without SR : code/pre_trained/model_withoutsr_pretrained.pt

• Model of eSL-Net with SR: code/model_sr.py

• Model of eSL-Net without SR: code/model_withoutsr.py

• Example files with event data: data_example

  • Data of APS frames: data_example/cups/images、 data_example/pic2/images

  • Data of events: data_example/cups/mat、 data_example/pic2/mat

    if you have new event data, you can preprocess the events referring to data_example/cups/mat and data_example/pic2/mat.

  • the path of loading input for eSL-Net: realdata_list.txt

• Run reconstruction:

  cd code
  

  if you want to reconstruct images without SR:

  python test_realdata.py --sr=0 --model=pre_trained/model_withoutsr_pretrained.pt --num_frame=3 --output_path=realdata_dn/
  

  if you want to reconstruct images with SR:

  python test_realdata.py --sr=1 --model=pre_trained/model_withsr_pretrained.pt --num_frame=3 --output_path=realdata_sr/
  

Synthetic Dataset

This synthetic dataset is generated from high-resolution sharp images of GoPro dataset and ESIM. And the process of generating the synthetic dataset is described in detailed in our paper.

Downloads are available via Baidu Net Disk.

Contents

• Introduction
• Results
  • Comparisons of Reconstruction on the synthetic dataset
  • Comparisons of Reconstruction on the real dataset
  • High frame-rate Reconstruction

Introduction

With extremely high temporal resolution, event cameras have a large potential for robotics and computer vision. However, the recovering of high-quality images from event cameras is a very challenge problem, where the following issues should be addressed simultaneously.

• Low frame-rate and blurry intensity images: The APS (Active Pixel Sensor) frames are with relatively low frame-rate. And the motion blur is inevitable when recording highly dynamic scenes.

• High level and mixed noises: The thermal effects or unstable light environment can produce a huge amount of noisy events. Together with the noises from APS frames, the reconstruction of intensity image would fall into a mixed noises problem.

• Low spatial-resolution: The leading commercial event cameras are typically with very low spatial-resolution. And there is a balance between the spatial-resolution and the latency

In our paper, we propose an explainable network, an event-enhanced Sparse Learning Network (eSL-Net), to recover the high-quality images from event cameras. Since events depict brightness changes, with the enhanced degeneration model by the events, the clear and sharp high-resolution latent images can be recovered from the noisy, blurry and low-resolution intensity observations. Exploiting the framework of sparse learning, the events and the low-resolution intensity observations can be jointly considered. Furthermore, without additional training process, the proposed eSL-Net can be easily extended to generate continuous frames with frame-rate as high as the events.

Results of Reconstruction

Qualitative Comparisons of Reconstruction on the synthetic dataset

Qualitative Comparisons of Reconstruction on the real dataset

High frame-rate Reconstruction

In the following videos, The left side is the original APS frame by bicubic upsampling for 4 times, and the right side are the high frame rate, high resolution reconstructed results of eSL-Net.

Event Camera——DAVIS240：

Event Camera——DAVIS346：