FrameCorr: Adaptive, Autoencoder-based Neural Compression for Video Reconstruction in Resource and Timing Constrained Network Settings

TL;DR

FrameCorr is a neural compression method that uses autoencoders and previous data to reconstruct video frames from incomplete data, addressing bandwidth and timing constraints in IoT video transmission.

Contribution

It introduces a novel autoencoder-based neural compression technique that predicts missing frame segments using prior data, improving video reconstruction under resource constraints.

Findings

Effective reconstruction of incomplete video frames

Improved robustness to network bandwidth limitations

Enhanced video quality in IoT scenarios

Abstract

Despite the growing adoption of video processing via Internet of Things (IoT) devices due to their cost-effectiveness, transmitting captured data to nearby servers poses challenges due to varying timing constraints and scarcity of network bandwidth. Existing video compression methods face difficulties in recovering compressed data when incomplete data is provided. Here, we introduce FrameCorr, a deep-learning based solution that utilizes previously received data to predict the missing segments of a frame, enabling the reconstruction of a frame from partially received data.