error-resilience

EE6427 Video Signal Processing

TensorFI 2 is a fault injection framework for injecting both hardware and software faults into applications written using the TensorFlow 2 framework.

Monitors MCP tool error rates and latencies, automatically tripping circuit breakers and routing requests to backup tools | GenPark: https://www.genpark.ai | LinkedIn: https://www.llinkedin.com/company/92955550

Monitors MCP tool error rates and latencies, automatically tripping circuit breakers and routing requests to backup tools | GenPark: https://www.genpark.ai | LinkedIn: https://www.llinkedin.com/company/92955550

This project is to study fault injection testing in which we used the TensorFI framework to evaluate PilotNet's faultolerance ability. PilotNet is an Nvidia's End to End Learning on Steering for Self-Driving Car model.

TensorFlow Data Mutator (TF-DM) is a framework for injecting different data faults into ML applications written using the TensorFlow 2 framework.

Testbench: can hardware H.264/H.265 decoders handle missing slices? Measured on Rockchip RK3588 (MPP) and Intel N100 (VA-API), with software fixes

A criticality-aware H.264 encoder simulation that models how different encoding blocks can be protected with resilience strategies such as retries, ECC, TMR, and checkpointing. This project demonstrates how fault-tolerant design principles can be applied to video compression pipelines.