Adity Gupta

Tech-savvy innovator with hands-on experience in emerging technologies and a passion for continuous improvement. Skilled in identifying opportunities for technological enhancements and implementing effective solutions. Adept at leveraging new tools and methods to solve problems and enhance productivity. Excels in adapting to fast-paced environments and driving technological advancements.

Human Action Recognition using CNN

Developed a deep learning model for Human Action Recognition using Convolutional Neural Networks (CNNs), achieving 91% accuracy on a specialized dataset. This group project involved preprocessing video data, extracting and augmenting frames, and training the model to classify various human activities. Optimized model performance through tuning and evaluation using TensorFlow, OpenCV, and Python.

Image Captioning using VGG16

Developed an Image Captioning model using VGG16 as an encoder and LSTM as a decoder to generate descriptive captions for images. Preprocessed image and text data, trained on image-caption pairs, and implemented inference using greedy and beam search. Achieved accurate, meaningful caption generation using TensorFlow/Keras, Python, and NLTK.