Script that receives an Image (original) and a set of images to be used as "pixels" in reconstruction of the Original image using the set of images as "pixels"

Last update: Oct 24, 2021

Related tags

Deep Learning picinpics

Overview

picinpics

Script that receives an Image (original) and a set of images to be used as "pixels" in reconstruction of the Original image using the set of images as "pixels"

Requirements

Python 3.8
Poetry
Pyenv [OPTIONAL]

Install

poetry install

Usage

poetry run python main.py

Execution log:

2021-10-24 15:47:55.390 | DEBUG    | __main__:<module>:107 - to grid image read
2021-10-24 15:47:56.226 | INFO     | __main__:<module>:111 - reading files
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 21/21 [00:00<00:00, 1081.52it/s]
2021-10-24 15:47:56.318 | INFO     | __main__:<module>:113 - 20 retrieved files
2021-10-24 15:47:56.319 | INFO     | __main__:<module>:115 - images normalizing
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:00<00:00, 458.17it/s]
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:00<00:00, 597.57it/s]
2021-10-24 15:47:56.399 | INFO     | __main__:<module>:119 - quantizing color
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:00<00:00, 857.52it/s]
2021-10-24 15:47:56.425 | INFO     | __main__:<module>:122 - building grid
2021-10-24 15:47:56.431 | INFO     | __main__:build_grid:82 - quantizing to grid image
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8160/8160 [00:06<00:00, 1257.16it/s]
2021-10-24 15:48:02.929 | INFO     | __main__:build_grid:85 - finding closests images
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8160/8160 [00:16<00:00, 500.46it/s]
2021-10-24 15:48:19.237 | INFO     | __main__:build_grid:92 - assemblying grid
2021-10-24 15:48:19.472 | INFO     | __main__:<module>:126 - done

Result

To increase the output quality change SUB_IMAGE_SIZE in code (W at image below), but this will dramatically increase the processing time.

Original	Grid Image

Algorithm

Improves

There is two major bottlenecks regarding processing:

Quantization

Receives an image as input and returns a single pixel 3D Tuple of UINT8
Match

Find the closest Quantized Pixel from Pixel Images to a given Quantized Pixel from Original Image

Script that receives an Image (original) and a set of images to be used as "pixels" in reconstruction of the Original image using the set of images as "pixels"

Related tags

Overview

picinpics

Requirements

Install

Usage

Result

Algorithm

Improves

Owner

RodrigoCMoraes

A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains (IJCV submission)

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

Deep Multi-Magnification Network for multi-class tissue segmentation of whole slide images

Dynamic Realtime Animation Control

[NeurIPS 2021] SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning

LEAP: Learning Articulated Occupancy of People

Ego4d dataset repository. Download the dataset, visualize, extract features & example usage of the dataset

CVPR 2021 - Official code repository for the paper: On Self-Contact and Human Pose.

Hyperparameters tuning and features selection are two common steps in every machine learning pipeline.

[CVPR 2022 Oral] Balanced MSE for Imbalanced Visual Regression https://arxiv.org/abs/2203.16427

Network Compression via Central Filter

CenterFace(size of 7.3MB) is a practical anchor-free face detection and alignment method for edge devices.

Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encoding

Detecting drunk people through thermal images using Deep Learning (CNN)

Code for the ICCV'21 paper "Context-aware Scene Graph Generation with Seq2Seq Transformers"

Simple-Image-Classification - Simple Image Classification Code (PyTorch)

KIDA: Knowledge Inheritance in Data Aggregation

Automatic tool focused on deriving metallicities of open clusters

PyTorch framework for Deep Learning research and development.

Code for TIP 2017 paper --- Illumination Decomposition for Photograph with Multiple Light Sources.