Bird Species Classification using Inception-v3 Network

Applications to identify birds based on their appearance and taxonomy

Determine the 🐦 from its 🎵

There are about 10,000 different bird species in the world, and they play an important role in the natural world. They serve as good indicators of declining habitat quality and pollution. It is often easier to hear birds than it is to see them. Bird_CLEF 2021 - Birdcall Identification is a Kaggle competition organized by The Cornell Lab of Ornit…

Explore deep learning-powered image classification with PyTorch. Achieved 98% accuracy on Natural Images and 95% on Birds Species using AlexNet and EfficientNet-B1. Dive into the code and results!

bird_classification web application

ResNet-34 Model trained from scratch to classify 450 different species of birds with 98.6% accuracy.

MVA - Kaggle Challenge - Bird Image Recognition

Fine-grained species classification

Selected Topics in Visual Recognition using Deep Learning, NYCU. CodaLab competition - Bird images classification

This project is a bird classifier that uses the PyTorch framework and the ResNet50 model. It can recognize the species of birds in images based on their visual features. It supports 200 different categories of birds.

Experiment testing the feasibility of individual bird recognition from audio recordings

Bird Sound Recognize

Signature Work @ DKU: Large Scale Bird Sound Recognition in China Region

Free open information (CC0) about nature on planet earth

Constructed an advanced deep learning model utilizing TensorFlow and CNN to precisely classify bird species from audio inputs. Leveraging Librosa for audio processing, the model is trained on extensive bird sound datasets to ensure robust performance. Integrated with Hugging Face, our application facilitates seamless uploading of bird sound samples

Code used for my final project in Computer Vision at Texas State University, Spring 2019

BirdNET as a systemd service with other features.

Source code for BMBF InnoTruck demo of BirdNET.