
English → Hindi Translation
From ‘Hello’ to ‘नमस्ते’ in milliseconds — this sequence-to-sequence with LSTM translation engine bridges English and Hindi with speed and precision.
From ‘Hello’ to ‘नमस्ते’ in milliseconds — this sequence-to-sequence with LSTM translation engine bridges English and Hindi with speed and precision.
Meet Harro: a Blender-designed robot performs real-time human pose estimation in a ROS2 Gazebo simulation environment. Bringing robotics and AI together.
Imagine a model that not only identifies but also pinpoints the location of everyday objects like an AC, fan, or stove! This powerful localization model offers precise object detection, enhancing everyday applications with advanced capabilities.
Explore a model that generates descriptive captions for images, bridging the gap between visual content and natural language. This project demonstrates how AI can understand and describe the content of an image.
My project features a cutting-edge Generative Adversarial Network (GAN) that generates realistic handwritten digits ranging from 0 to 9. Trained on the widely-used MNIST dataset.
I developed a categorical classification model using an own cnn architecture to detect different emotions of humans. The model was trained on FER dataset taken from kaggle.
I've developed a natural language processing model designed to classify movie reviews as either positive or negative. Trained on the renowned IMDB dataset, the model excels in accurately categorizing new reviews.
I developed a categorical classification model using an own cnn architecture to classify different types of foods. The model was trained on a diverse dataset of labeled images containing pizza, burger, ice cream, samosa and sushi.
I've built a robust categorical classification model specifically designed to recognize handwritten digits spanning from 0 to 9. Leveraging the well-known MNIST dataset, this model achieves remarkable accuracy.
I developed a binary classification model using a convolutional neural network to distinguish between parrots and sparrows. The model was trained on a diverse dataset of labeled images, achieving high accuracy.