There are mainly two types of super-resolution methods: traditional methods and deep learning methods. While traditional methods define closed-form expressions with assumptions, deep learning methods rely on priors learned from data sets. However, both of them have disadvantages such as being too simple and having strong trust in priors. We focus on how to generate a high-resolution image using low-resolution images without priors by utilizing spatial hash encoding. We propose a grid-based super-resolution model using spatial hash encoding to map coordinate information into higher dimensional space. Our aim is to eliminate long training times and not rely on priors from data sets that are not able to cover all real-world scenarios. Therefore, our proposed model is able to do task-specific super-resolution without priors and eliminate potential hallucination effects caused by wrong priors.
- Docker 24.0.5
- Docker Compose v2.20.2
Clone the repository.
git clone https://github.com/veliglmz/grid-based-super-resolution-using-spatial-hash-encoding.git
cd grid-based-super-resolution-using-spatial-hash-encoding
Build the docker image.
docker compose build
Run the docker image. (the outputs are in the results folder of the host.)
docker compose run app
Stop containers and remove containers, networks, volumes, and images.
docker compose down
We use NTIRE22 BURSTSR Dataset Generation framework with forward and inverse camera pipeline code from timothybrooks/unprocessing.
In the datasets/NTIRE22_BURSTSR_dataset_generation/main.py, there are two options: MFSR and SISR.
It creates a dataset from original images randomly based on types. MFRS for 4x upsamling and SISR for 8x upsampling.
cd datasets/NTIRE22_BURSTSR_dataset_generation
python main.py -t MFSR