Introduction

The Maya CLI is a 100% open source, contributor-friendly command line tool written in go to generate and verify zero-knowledge proof of image transformations.

Install

To run the commands, you can either use docker or build from source.

Docker

Install docker and run these commands using docker.

Build from source

1. Build and install the binary

   go install .
   

2. Test if it is installed properly

   maya-cli --help
   

Run maya

You can run maya commands easily using docker.

Maya CLI provides commands for both proving an image transformation and verifying it.

You can also look at the available transformations for proving by running:

docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove --help

Similarly, to look at the available transformations for verifying, run:

docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify --help

Crop

To run sample code for crop transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is cropped correctly, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove crop \
   --original-image=./sample/original.png \
   --final-image=./sample/cropped.png \
   --height-start-new=0 \
   --width-start-new=0 \
   --proof-dir=proofs
   

4. To verify that an image is cropped correctly, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify crop \
   --final-image=./sample/cropped.png \
   --proof-dir=proofs
   

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Rotate 90

To run sample code for rotate 90 transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is rotated correctly, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove rotate90 \
   --original-image=./sample/original.png \
   --final-image=./sample/rotate90.png \
   --proof-dir=proofs
   

4. To verify that an image is cropped correctly, run:

docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify rotate90 \
--final-image=./sample/rotated90.png \
--proof-dir=proofs

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Rotate 180

To run sample code for rotate 180 transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is rotated correctly, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove rotate180 \
   --original-image=./sample/original.png \
   --final-image=./sample/rotate180.png \
   --proof-dir=proofs
   

4. To verify that an image is cropped correctly, run:

docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify rotate180 \
--final-image=./sample/rotated180.png \
--proof-dir=proofs

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Rotate 270

To run sample code for rotate 270 transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is rotated correctly, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove rotate270 \
   --original-image=./sample/original.png \
   --final-image=./sample/rotate270.png \
   --proof-dir=proofs
   

4. To verify that an image is cropped correctly, run:

docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify rotate270 \
--final-image=./sample/rotate270.png \
--proof-dir=proofs

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Flip vertical

To run sample code for flip vertical transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is correctly flipped vertically, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove flip-vertical \
   --original-image=./sample/original.png \
   --final-image=./sample/flipped_vertical.png \
   --proof-dir=proofs
   

4. To verify that an image is correctly flipped vertically, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify flip-vertical \
   --final-image=./sample/flipped_vertical.png \
   --proof-dir=proofs
   

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Flip Horizontal

To run sample code for flip horizontal transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is correctly flipped horizontally, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove flip-horizontal \
   --original-image=./sample/original.png \
   --final-image=./sample/flipped_horizontal.png \
   --proof-dir=proofs
   

4. To verify that an image is correctly flipped horizontally, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify flip-horizontal \
   --final-image=./sample/flipped_horizontal.png \
   --proof-dir=proofs
   

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Brighten

To run sample code for crop transformation, follow these steps:

1. Clone the maya-cli repository

   git clone https://github.com/0xmayalabs/maya-cli.git
   

2. Cd into the directory

   cd maya-cli
   

3. To prove that an image is correctly brightened by a brightness factor factor, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest prove brighten \
   --original-image=./sample/original.png \
   --final-image=./sample/brightened.png \
   --brightening-factor=2 \
   --proof-dir=proofs
   

4. To verify that an image is correctly brightened by a brightness factor factor, run:

   docker run --rm -v "$(pwd):/opt/maya" 0xmayalabs/maya-cli:latest verify brighten \
   --final-image=./sample/brightened.png \
   --proof-dir=proofs
   

Please note that the repository contains sample images that you can use to get started quickly, but you don't need to clone the repository to run the prove or verify commands.

Performance

We evaluate the efficiency and performance of generating zero-knowledge (ZK) proofs for image transformations through the development of custom ZK circuits.

These circuits are developed using gnark, an open-source framework to design ZK circuits. Our objective is to benchmark the performance of our custom circuits against general purpose zkVMs such as RiscZero and SP1.

It's important to note that during this performance evaluation, the original image remains private, and only the transformed image is revealed.

Macbook Pro

The performance is measured on Macbook Pro M1, 10 cores of CPU & 16GB RAM.

Crop

Rotate 90

Rotate 180

Rotate 270

Flip vertical

Flip horizontal

AWS EC2 R6i.8xlarge

The performance is measured on AWS EC2 R6i.8xlarge instance, with 32 cores of CPU and 256 GB RAM.

Crop

Rotate 90

Rotate 180

Rotate 270

Flip vertical

Flip horizontal