machine-learning
an archive of projects with this tag
Energy-Aware Camera Scheduling for Opportunistic 3D Mapping: Wearable devices operate under stringent energy constraints and cannot continuously capture video. This raises a fundamental systems question — given a limited sensing budget, when should a mobile device collect visual data to maximize the coverage and freshness of a continuously evolving 3D map? We address this with IndoorMapper, a budget-aware sensing framework for opportunistic 3D map maintenance that localizes users within an existing map, estimates the utility of potential observations, and dynamically allocates a user-specified sensing budget across a user’s trajectory. |
BLINC: A Fault-Tolerant Neural Codec for Wireless Real-time Video Streaming: Wireless real-time video applications such as AR/VR, drones, cloud gaming, and broadcast streaming cannot always wait for retransmissions when packets are dropped or corrupted. In this project, we seek to understand whether a video codec preserve visual quality when wireless links deliver partially corrupted packets rather than only clean packets or full packet losses. We address this with BLINC, a bit-loss-insensitive neural codec that combines segmented entropy encoding, segment-wise loss-aware fine-tuning, and Wi-Fi link-layer integration to enable graceful degradation under bit-level corruption while reducing retransmissions and wireless airtime. |
Plug-and-Play Self-Supervised Temporal Consistency Refinement for Monocular Depth Estimation in Video: For the final project in MIT’s Introduction to Computer Vision class, I worked on a lightweight self-supervised refinement model with PyTorch that improves monocular video depth estimation by enforcing temporal and visual consistency across frames. Our plug-in approach enhances single-image depth predictions with smooth, stable results while remaining more efficient than bulky state-of-the-art methods. |
| RISC-V Processor Optimized for Neural Network Inference: Built CPU with 4-stage pipeline, bypassing, and custom instruction set extensions for inference. Implemented loop unrolling and function inlining code optimizations. |
Iterative Neural Network Based Approach to Automated IFT-20 Sensory Neuron Identification in Caenorhabditis elegans: Developed custom multistage neural network implemented using PyTorch to classify 22 sensory neurons in the IFT-20 subunit of the C. elegans brain. Paper available upon request |
Using Machine Learning to Augment Dynamic Time Warping Based Signal Classification: Designed neural network using TensorFlow that predicts likely distortions between signals to eliminate unnecessary computations in Dynamic Time Warping (DTW), a popular signal comparison algorithm. arXiv:2206.07200 |
Object Tracking & Gesture Recognition: Created interactive game with real-time hand-tracking and neural network-based gesture identification using OpenCV and TensorFlow in Python. 1st place term project. |
Enabling High-Accuracy Human Activity Recognition with Fine-Grained Indoor Localization: Used multilateration with fine-grained distance estimates from Wi-Fi Round Trip Time to perform high-accuracy indoor localization. Combined location, accelerometer, and audio data collected from a custom-made Android application programmed in Java to train a neural network that performs activity recognition of various household tasks. arXiv:2108.06838 |