Due to the limitation of shooting conditions in the real world, there exist many insufficient-resolution videos. To be transmitted under limited bandwidth conditions, low-resolution videos often have to be compressed further, which introduces more compression artifacts and results in severer damage to the video quality. Accordingly, video super-resolution in practical applications must repair compressed damage and down-sampling damage simultaneously, which can be intuitively solved by two subtasks, including compressed video quality enhancement (VQE) and video super-resolution (VSR). Therefore, we proposed a novel model, the Feature Multiplexing Video Super-Resolution for Compressed Video (FM-VSR) to effectively handle such multi-tasks. Firstly, we use a complicated Deformable Regression Pyramid (DRP) module to align the reference frame and each supporting frame. Then feature multiplexing structure is utilized in VSR to make better use of the VQE information. Finally, a skip-attention structure is introduced in the reconstruction module to predict the high-quality video frame. Many experiments on benchmark datasets show that our method can achieve better performance both in quantitative and qualitative than the state-of-the-art (SOTA) two-stage approaches.

In this paper, we propose a neural representation for videos that enables real-time quality-scalable decoding, called QS-NeRV. QS-NeRV comprises a Self-Learning Distribution Mapping Network (SDMN) and Extensible Enhancement Networks (EENs). Firstly, SDMN functions as the base layer (BL) for scalable video coding, focusing on encoding videos of lower quality. Within SDMN, we employ a methodology that minimizes the bitstream overhead to achieve efficient information exchange between the encoder and decoder instead of direct transmission. Specifically, we utilize an invertible network to map the multi-scale information obtained from the encoder to a specific distribution. Subsequently, during the decoding process, this information is recovered from a randomly sampled latent variable to assist the decoder in achieving improved reconstruction performance. Secondly, EENs serve as the enhancement layers (ELs) and are trained in an overfitting manner to obtain robust restoration capability. By integrating the fixed BL bitstream with the parameters of EEN as an extension pack, the decoder can produce higher-quality enhanced videos. Furthermore, the scalability of the method allows for adjusting the number of combined packs to accommodate diverse quality requirements. Experimental results demonstrate our proposed QS-NeRV outperforms the state-of-the-art real-time decoding INR-based methods on various datasets for video compression and interpolation tasks.

We herein propose an efficient video coding system (EVCS) consists of a conventional codec and an adaptive overfitted multi-scale attention network (MSAN) to improve coding efficiency. At the encoder side, the MSAN adjusts the network size adaptively and is trained in an overfitting way for a group of frames. Using only the current encoding video stream as a training set, the MSAN can easily obtain powerful restoration capability. After training, the learned parameters of the MSAN are transmitted to the decoder as part of the encoded bitstream. At the decoder side, the MSAN loaded with the transmitted parameters can restore the reconstructed frames very meticulously. Compared with the high efficiency video coding (HEVC) standard, the EVCS can achieve 12.141% Bjøntegaard-Delta bitrate reduction, which outperforms existing deep learning based compressed video restoration works with less computation complexity. Moreover, the MSAN is an additional part of the conventional codec without any structure change, thereby rendering it compatible with the existing coding systems.