Pmapper is a super-resolution and deconvolution toolkit for python 3.6+

Related tags

Deep Learningpmapper
Overview

pmapper

pmapper is a super-resolution and deconvolution toolkit for python 3.6+. PMAP stands for Poisson Maximum A-Posteriori, a highly flexible and adaptable algorithm for these problems. An implementation of the contemporary Richardson-Lucy algorithm is included for comparison.

The name of this repository is an homage to MTF-Mapper, a slanted edge MTF measurement program written by Frans van den Bergh.

The implementations of all algorithms in this repository are CPU/GPU agnostic and performant, able to perform 4K restoration at hundreds of iterations per second.

Usage

Basic PMAP, Multi-frame PMAP

import pmapper

img = ... # load an image somehow
psf = ... # acquire the PSF associated with the img
pmp = pmapper.PMAP(img, psf)  # "PMAP problem"
while pmp.iter < 100:  # number of iterations
    fHat = pmp.step()  # fHat is the object estimate

In simulation studies, the true object can be compared to fHat (for example, mean square error) to track convergence. If the psf is "larger" than the image, for example a 1024x1024 image and a 2048x2048 psf, the output will be super-resolved at the 2048x2048 resolution.

PMAP is able to combine multiple images of the same objec with different PSFs into one with the multi-frame variant. This can be used to combat dynamical atmospheric seeing conditions, line of sight jitter, or even perform incoherent aperture synthesis; rendering images from sparse aperture systems that mimic or exceed a system with a fully filled aperture.

import pmapper

# load a sequence of images; could be any iterable,
# or e.g. a kxmxn ndarray, with k = num frames
# psfs must have the same "size" (k) and correspond
# to the images in the same indices
imgs = ...
psfs = ...
pmp = pmapper.MFPMAP(imgs, psfs)  # "PMAP problem"
while pmp.iter < len(imgs)*100:  # number of iterations
    fHat = pmp.step()  # fHat is the object estimate

Multi-frame PMAP cycles through the images and PSFs, so the total number of iterations "should" be an integer multiple of the number of source images. In this way, each image is "visited" an equal number of times.

GPU computing

As mentioned previously, pmapper can be used trivially on a GPU. To do so, simply execute the following modification:

import pmapper
from pmapper import backend

import cupy as cp
from cupyx.scipy import (
    ndimage as cpndimage,
    fft as cpfft
)

backend.np._srcmodule = cp
backend.fft.fft = cpfft
backend.ndimage._srcmodule = cpndimage

# if your data is not on the GPU already
img = cp.array(img)
psf = cp.array(psf)

# ... do PMAP, it will run on a GPU without changing anything about your code

fHatCPU = fHat.get()

cupy is not the only way to do so; anything that quacks like numpy, scipy fft, and scipy ndimage can be used to substitute the backend. This can be done dynamically and at runtime. You likely will want to cast your imagery from fp64 to fp32 for performance scaling on the GPU.

Owner
NASA Jet Propulsion Laboratory
A world leader in the robotic exploration of space
NASA Jet Propulsion Laboratory
Exploring Cross-Image Pixel Contrast for Semantic Segmentation

Exploring Cross-Image Pixel Contrast for Semantic Segmentation Exploring Cross-Image Pixel Contrast for Semantic Segmentation, Wenguan Wang, Tianfei Z

Tianfei Zhou 510 Jan 02, 2023
Matthew Colbrook 1 Apr 08, 2022
Open source person re-identification library in python

Open-ReID Open-ReID is a lightweight library of person re-identification for research purpose. It aims to provide a uniform interface for different da

Tong Xiao 1.3k Jan 01, 2023
Checking fibonacci - Generating the Fibonacci sequence is a classic recursive problem

Fibonaaci Series Generating the Fibonacci sequence is a classic recursive proble

Moureen Caroline O 1 Feb 15, 2022
Tensorflow Implementation of Pixel Transposed Convolutional Networks (PixelTCN and PixelTCL)

Pixel Transposed Convolutional Networks Created by Hongyang Gao, Hao Yuan, Zhengyang Wang and Shuiwang Ji at Texas A&M University. Introduction Pixel

Hongyang Gao 95 Jul 24, 2022
An inofficial PyTorch implementation of PREDATOR based on KPConv.

PREDATOR: Registration of 3D Point Clouds with Low Overlap An inofficial PyTorch implementation of PREDATOR based on KPConv. The code has been tested

ZhuLifa 14 Aug 03, 2022
Pytorch implementation of RED-SDS (NeurIPS 2021).

Recurrent Explicit Duration Switching Dynamical Systems (RED-SDS) This repository contains a reference implementation of RED-SDS, a non-linear state s

Abdul Fatir 10 Dec 02, 2022
Pytorch implementation of our paper under review -- 1xN Pattern for Pruning Convolutional Neural Networks

1xN Pattern for Pruning Convolutional Neural Networks (paper) . This is Pytorch re-implementation of "1xN Pattern for Pruning Convolutional Neural Net

Mingbao Lin (林明宝) 29 Nov 29, 2022
Learning 3D Part Assembly from a Single Image

Learning 3D Part Assembly from a Single Image This repository contains a PyTorch implementation of the paper: Learning 3D Part Assembly from A Single

18 Dec 21, 2022
Transformer in Computer Vision

Transformer-in-Vision A paper list of some recent Transformer-based CV works. If you find some ignored papers, please open issues or pull requests. **

506 Dec 26, 2022
Implementation of a Transformer that Ponders, using the scheme from the PonderNet paper

Ponder(ing) Transformer Implementation of a Transformer that learns to adapt the number of computational steps it takes depending on the difficulty of

Phil Wang 65 Oct 04, 2022
Fine-grained Post-training for Improving Retrieval-based Dialogue Systems - NAACL 2021

Fine-grained Post-training for Multi-turn Response Selection Implements the model described in the following paper Fine-grained Post-training for Impr

Janghoon Han 83 Dec 20, 2022
Semantic segmentation models, datasets and losses implemented in PyTorch.

Semantic Segmentation in PyTorch Semantic Segmentation in PyTorch Requirements Main Features Models Datasets Losses Learning rate schedulers Data augm

Yassine 1.3k Jan 07, 2023
The official implementation code of "PlantStereo: A Stereo Matching Benchmark for Plant Surface Dense Reconstruction."

PlantStereo This is the official implementation code for the paper "PlantStereo: A Stereo Matching Benchmark for Plant Surface Dense Reconstruction".

Wang Qingyu 14 Nov 28, 2022
KwaiRec: A Fully-observed Dataset for Recommender Systems (Density: Almost 100%)

KuaiRec: A Fully-observed Dataset for Recommender Systems (Density: Almost 100%) KuaiRec is a real-world dataset collected from the recommendation log

Chongming GAO (高崇铭) 70 Dec 28, 2022
Large scale embeddings on a single machine.

Marius Marius is a system under active development for training embeddings for large-scale graphs on a single machine. Training on large scale graphs

Marius 107 Jan 03, 2023
[CVPR 2022] TransEditor: Transformer-Based Dual-Space GAN for Highly Controllable Facial Editing

TransEditor: Transformer-Based Dual-Space GAN for Highly Controllable Facial Editing (CVPR 2022) This repository provides the official PyTorch impleme

Billy XU 128 Jan 03, 2023
A PyTorch Toolbox for Face Recognition

FaceX-Zoo FaceX-Zoo is a PyTorch toolbox for face recognition. It provides a training module with various supervisory heads and backbones towards stat

JDAI-CV 1.6k Jan 06, 2023
A 10000+ hours dataset for Chinese speech recognition

WenetSpeech Official website | Paper A 10000+ Hours Multi-domain Chinese Corpus for Speech Recognition Download Please visit the official website, rea

310 Jan 03, 2023
ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation

ENet in Caffe Execution times and hardware requirements Network 1024x512 1280x720 Parameters Model size (fp32) ENet 20.4 ms 32.9 ms 0.36 M 1.5 MB SegN

Timo Sämann 561 Jan 04, 2023