Infinite-ISP

Infinite-ISP is a full-stack ISP development platform - from algorithm development to RTL design, FPGA/ASIC implementation, and associated firmware, tools, etc. It offers a unified platform that empowers ISP developers to accelerate ISP innovation. It includes a complete collection of camera pipeline modules written in Python, an FPGA bitstream & the associated Firmware for the implementation of the pipeline on the Kria KV260 development board, and lastly, a stand-alone Python-based Tuning Tool application for the pipeline. The main components of the Infinite-ISP project are listed below:

Repository name	Description
Infinite-ISP_AlgorithmDesign ⚓	Python based model of the Infinite-ISP pipeline for algorithm development
Infinite-ISP_ReferenceModel	Python based fixed-point model of the Infinite-ISP pipeline for hardware implementation
Infinite-ISP_FPGABinaries	FPGA binaries (bitstream + firmware executable) for the Xilinx® Kria KV260’s XCK26 Zynq UltraScale + MPSoC
Infinite-ISP_Firmware	Firmware for the Kria kV260’s embedded Arm® Cortex®A53 processor
Infinite-ISP_TuningTool	Collection of calibration and analysis tools for the Infinite-ISP

Infinite-ISP Algorithm Design: A Python-based Model for ISP Algorithm Development

Infinite-ISP Algorithm Design is a collections of camera pipeline modules implemented at the application level for converting an input RAW image from a sensor to an output RGB image. Infinite-isp aims to contain simple to complex algorithms at each modular level.

ISP pipeline for Infinite-ISP v1.1

Objectives

Many open-source ISPs are available over the internet. Most of them are developed by individual contributors, each having its strengths. This project aims to centralize all the open-source ISP development to a single place enabling all the ISP developers to have a single platform to contribute. InfiniteISP will not only contain the conventional algorithms but aims to contain state-of-the-art deep learning algorithms as well, enabling a clean comparison between the two. This project has no bounds to ideas and is aimed to include any algorithm that improves the overall results of the pipeline regardless of their complexity.

Feature Comparison Matrix

A comparison of features with the famous openISP.

InfiniteISP also tries to simulate the 3A-Algorithms.

Modules	infiniteISP	openISP
Crop	Bayer pattern safe cropping	----
Dead Pixel Correction	Modified Yongji's et al, Dynamic Defective Pixel Correction for Image Sensor	Yes
Black Level Correction	Calibration / sensor dependent - Applies BLC from config	Yes
Optical Electronic Transfer Function (OECF)	Calibration / sensor dependent - Implements a LUT from config	----
Anti Aliasing Filter	----	Yes
Digital Gain	Gains from config file	Brightness contrast control
Lens Shading Correction	To Be Implemented	----
Bayer Noise Reduction	Green Channel Guiding Denoising by Tan et al	Chroma noise filtering
White Balance	WB gains from config file	Yes
CFA Interpolation	Malwar He Cutler’s demosaicing algo	Yes - Malvar He Cutler
3A - Algorithms	AE & AWB	----
Auto White Balance	- Grey World - Norm 2 - PCA algorithm	----
Auto Exposure	- Auto Exposure based on skewness	----
Color Correction Matrix	Calibration / sensor dependent - 3x3 CCM from config	Yes - 4x3 CCM
Gamma Tone Mapping	Gamma LUT in RGB from config file	Yes - YUV and RGB domain
Color Space Conversion	YCbCr digital - BT 601 - Bt 709	Yes - YUV analogue
Contrast Enhancement	Modified contrast limited adaptive histogram equalization	----
Edge Enhancement / Sharpeining	Simple unsharp masking with strength control	Yes
Noise Reduction	Non-local means filter	Yes - NLM filter - Bilateral noise filter
Hue Saturation Control	----	Yes
RGB Conversion	Apply inverse conversion from YUV to RGB - same standard as CSC	No
Scale	- Integer Scaling - Non-Integer Scaling	----
False Color Suppression	----	Yes
YUV Format	- YUV - 444 - YUV - 422	----

Dependencies

The project is compatible with Python_3.9.12

The dependencies are listed in the requirements.txt file.

The project assumes pip package manager as a pre-requisite.

How to Run

Follow the following steps to run the pipeline

1. Clone the repo using

git clone https://github.com/10xEngineersTech/Infinite-ISP_ReferenceModel

2. Install all the requirements from the requirements file by running

pip install -r requirements.txt

3. Run isp_pipeline.py

python isp_pipeline.py

Example

There are a few sample images with tuned configurations already added to the project at in_frames/normal folder. In order to run any of these, just replace the config file name with any one of the sample configurations provided. For example to run the pipeline on Indoor1_2592x1536_12bit_RGGB.raw simply replace the config file name and data path in isp_pipeline.py

CONFIG_PATH = './config/Indoor1_2592x1536_12bit_RGGB-configs.yml'
RAW_DATA = './in_frames/normal/data'

How to Run on Pipeline on Multiple Images/Dataset

There is another script isp_pipeline_multiple_images.py that runs Infinite-ISP on multiple images with two modes:

1. DATASET PROCESSING
   Execute multiple images. Raw image should have its own config file with name <filename>-configs.yml where <filename> is raw filename otherwise the default configuration file configs.yml is used.

   For raw image format such as, NEF, DNG and CR2 we have also provided a funcationality to extract sensor information provided in these raw files metadata and update default config file.

2. VIDEO MODE
   Each image in the dataset is considered as video frame in sequence. All images use the same configuration parameters from configs.yml and 3A Stats calculated on a frame are applied to the next frame.

After cloning the repository and installing all the dependencies follow the following steps:

1. Set DATASET_PATH to dataset folder. For example if images are in in in_frames/normal/data folder

DATASET_PATH = './in_frames/normal/data'

2. If your dataset is present on another git repository you can use it as a submodule by using the following commands in the root directory. In the command, <url> is the address of git repository such as https://github.com/<user>/<repository_name and <path> is the location in your repository where you want to add the submodule and for Infinite ISP <path> should be ./in_frames/normal/<dataset_name>. Please keep in mind that your <dataset_name> should not be data because directory in_frames/normal/data already exists.

git submodule add <url> <path>
git submodule update --init --recursive

4. After adding git repository as a submodule update DATASET_PATH variable in isp_pipeline_dataset.py to ./in_frames/normal/<dataset_name>. Git does not allow to import a repository’s subfolder using a submodule. You can only add an entire repository and then access the folder. If you want to use images from a subfolder of a submodule modify the DATASET_PATH variable in isp_pipeline_dataset.py or video_processing.py accordingly.

DATASET_PATH = './in_frames/normal/<dataset_name>'

5. Run isp_pipeline_dataset.py or video_processing.py
6. The processed images are saved in out_frames folder.

Test Vector Generation

Please refer to the provided instructions for generating test vectors for multiple images, considering individual or multiple modules as the Device Under Test (DUT).

Contributing

Please read the Contribution Guidelines before making a Pull Request

Results

Here are the results of this pipeline compared with a market competitive ISP. The outputs of our ISP are displayed on the right, with the underlying ground truths on the left.

           ground truths                            infiniteISP

A comparison of the above results based on PSNR and SSIM image quality metrics

Images	PSNR	SSIM
Indoor1	20.0974	0.8599
Outdoor1	21.8669	0.9277
Outdoor2	20.3430	0.8384
Outdoor3	19.3627	0.8027
Outdoor4	20.7741	0.8561

User Guide

You can run the project by simply executing the isp_pipeline.py. This is the main file that loads all the algorithic parameters from the configs.yml The config file contains tags for each module implemented in the pipeline. A brief description as well as usage of each module is as follows:

Platform

platform	Details
filename	Specifies the file name for running the pipeline. The file should be placed in the in_frames/normal directory
disable_progress_bar	Enables or disables the progress bar for time taking modules
leave_pbar_string	Hides or unhides the progress bar upon completion

Sensor_info

sensor Info	Details
bayer_pattern	Specifies the bayer patter of the RAW image in lowercase letters - bggr - rgbg - rggb - grbg
range	Not used
bit_depth	The bit depth of the raw image
width	The width of the input raw image
height	The height of the input raw image
hdr	Not used

Crop

crop	Details
is_enable	Enables or disables this module. When enabled it ony crops if bayer pattern is not disturbed
is_debug	Flag to output module debug logs
new_width	New width of the input RAW image after cropping
new_height	New height of the input RAW image after cropping

Dead Pixel Correction

dead_pixel_correction	Details
is_enable	Enables or disables this module
is_debug	Flag to output module debug logs
dp_threshold	The threshold for tuning the dpc module. The lower the threshold more are the chances of pixels being detected as dead and hence corrected

HDR Stitching

To be implemented

Black Level Correction

black_level_correction	Details
is_enable	Enables or disables this module
r_offset	Red channel offset
gr_offset	Gr channel offset
gb_offset	Gb channel offset
b_offset	Blue channel offset
is_linear	Enables or disables linearization. When enabled the BLC offset maps to zero and saturation maps to the highest possible bit range given by the user
r_sat	Red channel saturation level
gr_sat	Gr channel saturation level
gb_sat	Gb channel saturation level
b_sat	Blue channel saturation level

Opto-Electronic Conversion Function

OECF	Details
is_enable	Enables or disables this module
r_lut	The look up table for oecf curve. This curve is mostly sensor dependent and is found by calibration using some standard technique

Digital Gain

digital_gain	Details
is_enable	This is a essential module and cannot be disabled
is_debug	Flag to output module debug logs
gain_array	Gains array. User can select any one of the gain listed here. This module works together with AE module
current_gain	Index for the current gain starting from zero

Lens Shading Calibration

To be implemented

Bayer Noise Reduction

bayer_noise_reduction	Details
is_enable	When enabled reduces the noise in bayer domain using the user given parameters
filt_window	Should be an odd window size
r_std_dev_s	Red channel gaussian kernel strength. The more the strength the more the blurring. Cannot be zero
r_std_dev_r	Red channel range kernel strength. The more the strength the more the edges are preserved. Cannot be zero
g_std_dev_s	Gr and Gb gaussian kernel strength
g_std_dev_r	Gr and Gb range kernel strength
b_std_dev_s	Blue channel gaussian kernel strength
b_std_dev_r	Blue channel range kernel strength

White balance

white_balance	Details
is_enable	Applies user given white balance gains when enabled
is_auto	When true enables the 3A - AWB and does not use the user given WB gains
r_gain	Red channel gain
b_gain	Blue channel gain

-->

3A - Auto White Balance (AWB)

auto_white_balance	Details
is_debug	Flag to output module debug logs
underexposed_pecentage	Set % of dark pixels to exclude before AWB gain calculation
overexposed_pecentage	Set % of saturated pixels to exclude before AWB gain calculation
algorithm	Can select one of the following algos - grey_world - norm_2 - pca
percentage	[0 - 100] - Parameter to select dark-light pixels percentage for pca algorithm

Color Correction Matrix (CCM)

color_correction_matrix	Details
is_enable	When enabled applies the user given 3x3 CCM to the 3D RGB image having rows sum to 1 convention
corrected_red	Row 1 of CCM
corrected_green	Row 2 of CCM
corrected_blue	Row 3 of CCM

Gamma Correction

gamma_correction	Details
is_enable	When enabled applies tone mapping gamma using the LUT
gamma_lut_8	The look up table for 8-bit gamma curve
gamma_lut_10	The look up table for 10-bit gamma curve
gamma_lut_12	The look up table for 12-bit gamma curve
gamma_lut_14	The look up table for 14-bit gamma curve

3A - Auto Exposure

auto_exposure	Details
is_enable	When enabled applies the 3A- Auto Exposure algorithm
is_debug	Flag to output module debug logs
center_illuminance	The value of center illuminance for skewness calculation ranges from 0 to 255. Default is 90
histogram_skewness	The range of histogram skewness should be between 0 and 1 for correct exposure calculation

Color Space Conversion (CSC)

color_space_conversion	Details
is_enable	This is a essential module and cannot be disabled
conv_standard	The standard to be used for conversion - 1 : Bt.709 HD - 2 : Bt.601/407

Contrast Enchancement

ldci	Details
is_enable	When enabled local dynamic contrast enhancement is applied to the Y channel
clip_limit	The clipping limit that controls amount of detail to be enhanced
wind	Window size for applying filter

Edge Enchancement / Sharpening

Sharpening	Details
is_enable	When enabled applies the sharpening
sharpen_sigma	Define the Standard Deviation of the Gaussian Filter
sharpen_strength	Controls the sharpen strength applied on the high frequency components

2d Noise Reduction

2d_noise_reduction	Details
is_enable	When enabled applies the 2D noise reduction
algorithm	Can select one of the following algos - nlm - ebf
window_size	Search window size for applying non-local means
patch_size	Patch size for applying mean filter
wts	Smoothening strength parameter
wind	Window size for applying entropy based bilateral filter
sigma	Range and spatial kernel parameter for entropy based bilateral filter

Scaling

scale	Details
is_enable	When enabled down scales the input image
is_debug	Flag to output module debug logs
new_width	Down scaled width of the output image
new_height	Down scaled height of the output image
is_hardware	When true applies the hardware friendly techniques for downscaling. This can only be applied to any one of the input sizes 3 input sizes and can downscale to - 2592x1944 to 1920x1080 or 1280x960 or 1280x720 or 640x480 or 640x360 - 2592x1536 to 1280x720 or 640x480 or 640x360 - 1920x1080 to to 1280x720 or 640x480 or 640x360
algorithm	Software friendly scaling. Only used when isHardware is disabled - Nearest_Neighbor - Bilinear
upscale_method	Used only when isHardware enabled. Upscaling method, can be one of the above algos
downscale_method	Used only when isHardware enabled. Downscaling method, can be one of the above algos

YUV Format

yuv_conversion_format	Details
is_enable	Enables or disables this module
conv_type	Can convert the YCbCr to YUV - 444 - 422

FAQ

Why is it named infiniteISP?

ISPs are hardware dependent. In them algorithms are limited to perform to their best because of hardware limitations. InfiniteISP tends to somewhat remove this limitation and let the algorithms perform to the full potential targeting best results.

Will inifniteISP also contain algorithms that involve machine learning?

Yes definitely this is mainly because it is seen that machine learning models tend to give perform much better results as compared to conventional models. The plan is as follows

• The release v0.x till v1.0 will involve buildng a basic ISP pipelne at conventional level.

• The release v1.0 will have all camera pipeline modules implemented at conventional level. This release will mostly contain algorithms that can be easily ported to hardware ISPs

• v1.x.x releases will have all the necessary improvements of these conventional algorithms till release v2.0

• From release v2.0 infiniteISP will start implementing machine learning models for specific algorithms.

• Release v3.0 will have infiniteISP having both conventional and deep learning algorithms (not for all pipeline modules but for specific ones)

License

This project is licensed under Apache 2.0 (see LICENSE file).

Acknowledgments

• This project started of from the inspiration of cruxopen/openISP

List of Open Source ISPs

• openISP
• Fast Open Image Signal Processor
• AbdoKamel - simple-camera-pipeline
• Mushfiqulalam - isp
• Karaimer - A Software Platform for Manipulating the Camera Imaging Pipeline
• rawpy