A video scene detection algorithm is designed to detect a variety of different scenes within a video

Last update: Jan 04, 2022

Overview

Scene-Change-Detection

The detection of scenes change is a simple problem that human beings face, but it gets much harder to handle autonomously a device that generally includes complex calculations and algorithms.

A video scene detection algorithm is designed to detect a variety of different scenes within a video. There is a very simple definition for a scene: It is a series of logically and chronologically related shots taken in a specific order to depict an over-arching concept or story. The identification of video scenes, in many video analysis applications, is a crucial pre-processing step. A dataset for video scene detection known as the Open Video Scene Detection (OVSD) dataset has been provided in order to evaluate algorithms for video scene detection. Videos in the dataset have an open-source nature, which makes them an ideal product to be used by academics, as well as industry researchers alike.

DATASET

Dataset 2012 - In the dataset, there are six video categories, and in each category, there are four to six video sequences
IBM video Scene Change Detection
A dataset for video scene detection known as the Open Video Scene Detection (OVSD) dataset has been provided in order to evaluate algorithms for video scene detection.

MODEL

VGG16 was used for this project.VGG16 is a convolutional neural network model proposed by K. Simonyan and A. Zisserman from the University of Oxford in the paper “Very Deep Convolutional Networks for Large- Scale Image Recognition”. The model achieves 92.7% top-5 test accuracy in ImageNet, which is a dataset of over 14 million images belonging to 1000 classes.

LIBRARIES USED

Numpy for array manipulation
OpenCV (cv2) for Image Augmentation
Keras for building the Neural Network
Matplotlib for plotting visuals

COMPILATION USED

Loss function selected is sparse categorical cross-entropy
Optimizer selected is Adam
Validation metric chosen is accuracy

Training

No of epochs = 5
Batch size = 1

REPORT

https://drive.google.com/file/d/1cwoP5cRJ5D76PvHV_WjCDRoSJIf0h9du/view?usp=sharing

COLLABORATORS

Neel kumar arya and Ashish Vidyarthi

A video scene detection algorithm is designed to detect a variety of different scenes within a video

Related tags

Overview

Scene-Change-Detection

DATASET

MODEL

LIBRARIES USED

COMPILATION USED

Training

REPORT

COLLABORATORS

License

Owner

IRON Kaggle project done while doing IRONHACK Bootcamp where we had to analyze and use a Machine Learning Project to predict future sales

For visualizing the dair-v2x-i dataset

text_recognition_toolbox: The reimplementation of a series of classical scene text recognition papers with Pytorch in a uniform way.

Code and Resources for the Transformer Encoder Reasoning Network (TERN)

The official PyTorch implementation of paper BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition

Code for our work "Activation to Saliency: Forming High-Quality Labels for Unsupervised Salient Object Detection".

This is an official implementation for "Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation".

Official repository for the paper "Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks"

The codes and related files to reproduce the results for Image Similarity Challenge Track 1.

Emotional conditioned music generation using transformer-based model.

Official pytorch implementation of DeformSyncNet: Deformation Transfer via Synchronized Shape Deformation Spaces

Robust, modular and efficient implementation of advanced Hamiltonian Monte Carlo algorithms

Pretraining on Dynamic Graph Neural Networks

Exemplo de implementação do padrão circuit breaker em python

PyTorch implementation of MLP-Mixer

Source code for CAST - Crisis Domain Adaptation Using Sequence-to-sequence Transformers (Accepted to ISCRAM 2021, CorePaper).

PyTorch code accompanying the paper "Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learning" (NeurIPS 2021).

Text to image synthesis using thought vectors

A Repository of Community-Driven Natural Instructions

Unofficial Pytorch Lightning implementation of Contrastive Syn-to-Real Generalization (ICLR, 2021)