✂️ EyeLipCropper is a Python tool to crop eyes and mouth ROIs of the given video.

Overview

EyeLipCropper

EyeLipCropper is a Python tool to crop eyes and mouth ROIs of the given video. The whole process consists of three parts: frame extraction, face alignment, and eye/mouth cropping. The cropped eye/mouth image size can be customized.

vis

Usage

Prerequisites

>>> pip install -r requirements.txt

1. Extract frames of a given video

>>> python frame_extract.py -h
usage: frame_extract.py [-h] [--video-path VIDEO_PATH] [--images-path IMAGES_PATH]

extract frames with opencv

optional arguments:
  -h, --help            show this help message and exit
  --video-path VIDEO_PATH
                        the input video path
  --images-path IMAGES_PATH
                        the output frames path
 
# default for test: this will generate frames of the video in `./test/images`
>>> python frame_extract.py

2. Align faces of the frames, with library face-alignment

>>> python face_align.py -h
usage: face_align.py [-h] [--images-path IMAGES_PATH] [--landmarks-path LANDMARKS_PATH] [--boxes-path BOXES_PATH] [--device DEVICE] [--log-path LOG_PATH]

align faces with `https://github.com/1adrianb/face-alignment`

optional arguments:
  -h, --help            show this help message and exit
  --images-path IMAGES_PATH
                        the input frames path
  --landmarks-path LANDMARKS_PATH
                        the output 68 landmarks path
  --boxes-path BOXES_PATH
                        the output bounding boxes path
  --device DEVICE       cpu or gpu cuda device
  --log-path LOG_PATH   logging when there are no faces detected
  
# default for test: this will generate landmarks and bounding boxes in
# `./test/landmarks` and `./test/boxes`
>>> python face_align.py

3. Crop left eye, right eye, mouth ROIs, with code modified from processing tools of [Eye] RT-GENE and [Mouth] LipForensics

>>> python eye_mouth_crop.py -h
usage: eye_mouth_crop.py [-h] [--images-path IMAGES_PATH] [--landmarks-path LANDMARKS_PATH] [--boxes-path BOXES_PATH] [--eye-width EYE_WIDTH] [--eye-height EYE_HEIGHT]
                         [--face-roi-width FACE_ROI_WIDTH] [--face-roi-height FACE_ROI_HEIGHT] [--left-eye-path LEFT_EYE_PATH] [--right-eye-path RIGHT_EYE_PATH]
                         [--mean-face MEAN_FACE] [--mouth-width MOUTH_WIDTH] [--mouth-height MOUTH_HEIGHT] [--start-idx START_IDX] [--stop-idx STOP_IDX]
                         [--window-margin WINDOW_MARGIN] [--mouth-path MOUTH_PATH]

crop eye and mouth regions

optional arguments:
  -h, --help            show this help message and exit
  --images-path IMAGES_PATH
                        [COMMON] the input frames path
  --landmarks-path LANDMARKS_PATH
                        [COMMON] the input 68 landmarks path
  --boxes-path BOXES_PATH
                        [EYE] the input bounding boxes path
  --eye-width EYE_WIDTH
                        [EYE] width of cropped eye ROIs
  --eye-height EYE_HEIGHT
                        [EYE] height of cropped eye ROIs
  --face-roi-width FACE_ROI_WIDTH
                        [EYE] maximize this argument until there is a warning message
  --face-roi-height FACE_ROI_HEIGHT
                        [EYE] maximize this argument until there is a warning message
  --left-eye-path LEFT_EYE_PATH
                        [EYE] the output left eye images path
  --right-eye-path RIGHT_EYE_PATH
                        [EYE] the output right eye images path
  --mean-face MEAN_FACE
                        [MOUTH] mean face pathname
  --mouth-width MOUTH_WIDTH
                        [MOUTH] width of cropped mouth ROIs
  --mouth-height MOUTH_HEIGHT
                        [MOUTH] height of cropped mouth ROIs
  --start-idx START_IDX
                        [MOUTH] start of landmark index for mouth
  --stop-idx STOP_IDX   [MOUTH] end of landmark index for mouth
  --window-margin WINDOW_MARGIN
                        [MOUTH] window margin for smoothed_landmarks
  --mouth-path MOUTH_PATH
                        [MOUTH] the output mouth images path

# default for test: this will generate the final cropped left eye,
# right eye, and mouth images in `./test/left_eye`, `./test/right_eye`
# , and `./test/mouth`
>>> python eye_mouth_crop.py
  • Note that the argument --face-roi-width and --face-roi-height should be maximized until there is a printed warning.

License

GPL-3.0 License

Reference

[1] Bulat, Adrian, and Georgios Tzimiropoulos. "How far are we from solving the 2d & 3d face alignment problem?(and a dataset of 230,000 3d facial landmarks)." Proceedings of the IEEE International Conference on Computer Vision (ICCV). 2017. GitHub: https://github.com/1adrianb/face-alignment

[2] Fischer, Tobias, Hyung Jin Chang, and Yiannis Demiris. "Rt-gene: Real-time eye gaze estimation in natural environments." Proceedings of the European Conference on Computer Vision (ECCV). 2018. GitHub: https://github.com/Tobias-Fischer/rt_gene

[3] Haliassos, Alexandros, et al. "Lips Don't Lie: A Generalisable and Robust Approach To Face Forgery Detection." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021. GitHub: https://github.com/ahaliassos/LipForensics/

Owner
Zi-Han Liu
Senior @ SJTU
Zi-Han Liu
Implements Gradient Centralization and allows it to use as a Python package in TensorFlow

Gradient Centralization TensorFlow This Python package implements Gradient Centralization in TensorFlow, a simple and effective optimization technique

Rishit Dagli 101 Nov 01, 2022
An Approach to Explore Logistic Regression Models

User-centered Regression An Approach to Explore Logistic Regression Models This tool applies the potential of Attribute-RadViz in identifying correlat

0 Nov 12, 2021
Source codes for "Structure-Aware Abstractive Conversation Summarization via Discourse and Action Graphs"

Structure-Aware-BART This repo contains codes for the following paper: Jiaao Chen, Diyi Yang:Structure-Aware Abstractive Conversation Summarization vi

GT-SALT 56 Dec 08, 2022
nfelo: a power ranking, prediction, and betting model for the NFL

nfelo nfelo is a power ranking, prediction, and betting model for the NFL. Nfelo take's 538's Elo framework and further adapts it for the NFL, hence t

6 Nov 22, 2022
KSAI Lite is a deep learning inference framework of kingsoft, based on tensorflow lite

KSAI Lite is a deep learning inference framework of kingsoft, based on tensorflow lite

80 Dec 27, 2022
DaReCzech is a dataset for text relevance ranking in Czech

Dataset DaReCzech is a dataset for text relevance ranking in Czech. The dataset consists of more than 1.6M annotated query-documents pairs,

Seznam.cz a.s. 8 Jul 26, 2022
Distributed Asynchronous Hyperparameter Optimization in Python

Hyperopt: Distributed Hyperparameter Optimization Hyperopt is a Python library for serial and parallel optimization over awkward search spaces, which

6.5k Jan 01, 2023
A chemical analysis of lipophilicities & molecule drawings including ML

A chemical analysis of lipophilicity & molecule drawings including a bit of ML analysis. This is a simple project that includes two Jupyter files (one

Aurimas A. Nausėdas 7 Nov 22, 2022
Official PyTorch code for the paper: "Point-Based Modeling of Human Clothing" (ICCV 2021)

Point-Based Modeling of Human Clothing Paper | Project page | Video This is an official PyTorch code repository of the paper "Point-Based Modeling of

Visual Understanding Lab @ Samsung AI Center Moscow 64 Nov 22, 2022
Serverless proxy for Spark cluster

Hydrosphere Mist Hydrosphere Mist is a serverless proxy for Spark cluster. Mist provides a new functional programming framework and deployment model f

hydrosphere.io 317 Dec 01, 2022
StyleGAN2-ADA-training-jupyter - Training custom datasets in styleGAN2-ADA by NVIDIA using Jupyter

styleGAN2-ADA-training-jupyter Training custom datasets in styleGAN2-ADA on Jupyter Official StyleGAN2-ADA by NIVIDIA Paper Training Generative Advers

Mang Su Hyun 2 Feb 24, 2022
Music Classification: Beyond Supervised Learning, Towards Real-world Applications

Music Classification: Beyond Supervised Learning, Towards Real-world Applications

104 Dec 15, 2022
SeqTR: A Simple yet Universal Network for Visual Grounding

SeqTR This is the official implementation of SeqTR: A Simple yet Universal Network for Visual Grounding, which simplifies and unifies the modelling fo

seanZhuh 76 Dec 24, 2022
Parameter-ensemble-differential-evolution - Shows how to do parameter ensembling using differential evolution.

Ensembling parameters with differential evolution This repository shows how to ensemble parameters of two trained neural networks using differential e

Sayak Paul 9 May 04, 2022
A Transformer-Based Siamese Network for Change Detection

ChangeFormer: A Transformer-Based Siamese Network for Change Detection (Under review at IGARSS-2022) Wele Gedara Chaminda Bandara, Vishal M. Patel Her

Wele Gedara Chaminda Bandara 214 Dec 29, 2022
This is the official implementation code repository of Underwater Light Field Retention : Neural Rendering for Underwater Imaging (Accepted by CVPR Workshop2022 NTIRE)

Underwater Light Field Retention : Neural Rendering for Underwater Imaging (UWNR) (Accepted by CVPR Workshop2022 NTIRE) Authors: Tian Ye†, Sixiang Che

jmucsx 17 Dec 14, 2022
A simple pygame dino game which can also be trained and played by a NEAT KI

Dino Game AI Game The game itself was developed with the Pygame module pip install pygame You can also play it yourself by making the dino jump with t

Kilian Kier 7 Dec 05, 2022
MIRACLE (Missing data Imputation Refinement And Causal LEarning)

MIRACLE (Missing data Imputation Refinement And Causal LEarning) Code Author: Trent Kyono This repository contains the code used for the "MIRACLE: Cau

van_der_Schaar \LAB 15 Dec 29, 2022
Fast and Context-Aware Framework for Space-Time Video Super-Resolution (VCIP 2021)

Fast and Context-Aware Framework for Space-Time Video Super-Resolution Preparation Dependencies PyTorch 1.2.0 CUDA 10.0 DCNv2 cd model/DCNv2 bash make

Xueheng Zhang 1 Mar 29, 2022
An easier way to build neural search on the cloud

An easier way to build neural search on the cloud Jina is a deep learning-powered search framework for building cross-/multi-modal search systems (e.g

Jina AI 17k Jan 02, 2023