LRConv-NeRV: Low Rank Convolution for Efficient Neural Video Compression

ArXi:2603.18261v1 Announce Type: cross Neural Representations for Videos (NeRV) encode entire video sequences within neural network parameters, offering an alternative paradigm to conventional video codecs. However, the convolutional decoder of NeRV remains computationally expensive and memory intensive, limiting its deployment in resource-constrained environments. This paper proposes LRCon-NeRV, an efficient NeRV variant that replaces selected dense 3x3 convolutional layers with structured low-rank separable convolutions, trained end-to-end within the decoder architecture.