Objective: Detect face-swap deepfake videos using a multimodal approach.
Techniques:
• CNNs for spatial feature extraction
• RNNs (LSTMs) for temporal sequence modeling
• GANs to ensure adversarial robustness
My Role:
• Dataset collection and video preprocessing
• Performed video scaling, normalization, face/landmark detection, frame extraction, and data augmentation
Technologies: Python, TensorFlow, Keras, OpenCV, CNN, LSTM, GAN
View on GitHubObjective: Create a voice-enabled assistant capable of automation and intelligent responses.
Features:
• Speech recognition
• Sentiment analysis
• Wikipedia summarization
• Movie recommendations
• NLP-based question answering system
Technologies: Python, NLTK, SpaCy, TensorFlow, SpeechRecognition API, Tkinter, PyAudio
View on GitHubObjective: Design a desktop application for managing hotel bookings and customers.
Highlights:
• GUI for room availability, reservations, check-ins, and check-outs
• MySQL integration with CRUD functionality
Technologies: Python (Tkinter), MySQL, mysql-connector
View on GitHubObjective: Analyze and classify user comments into Positive, Negative, or Neutral sentiments.
Work Done:
• Data preprocessing and text tokenization
• Sentiment classification using VADER
• Visualized results using Matplotlib and Seaborn
Tools: Python, NLTK, Pandas, VADER, Matplotlib, Seaborn
View on GitHub