This paper proposes a modified video compression model that optimizes vector quantization codebook by using the adapted Quantum Genetic Algorithm (QGA) that uses the quantum features, superposition, and entanglement to build optimal codebook for vector quantization. A context-based initial codebook is created by using a background subtraction algorithm; then, the QGA is adapted to get the optimal codebook. This optimal feature vector is then utilized as an activation function inside the neural network’s hidden layer to remove redundancy. Furthermore, approximation wavelet coefficients were lossless compressed with Differential Pulse Code Modulation (DPCM); whereas details coefficients are lossy compressed using Learning Vector Quantization (LVQ) neural networks. Finally, Run Length Encoding is engaged to encode the quantized coefficients to achieve a high compression ratio. As individuals in the QGA are actually the superposition of multiple individuals, it is less likely that good individuals will be lost. Experiments have proven the system’s ability to achieve a higher compression ratio with acceptable efficiency measured by PSNR.