In this work, we propose a deep learning-based adaptive quantization method to promote video coding performance. Due to inter-prediction and reference mechanism, the block-level quantization parameter (QP) not only influences current block distortion but also has complex temporal propagation effects on subsequent coding frames. Our idea is to utilize a deep network to model the complex temporal propagation relationship of quantization. As shown in Fig. 1, the deep network directly predicts all block-level QPs of the frame for the traditional encoder without changing the standard decoder. Since our network deploys only on the encoder side and has low inference complexity, it can be easily applied in practice. In addition, we use a learned coding network as a proxy of the traditional codec to train our network.

Video coding achieves efficient compression by exploiting the spatial and temporal correlations within the video signal. However, the noise in the source signal corrupts such correlation and impairs the coding performance. Motion compensated temporal filter (MCTF) [1] is a pre-processing tool that removes certain noise from the source signal, thereby enhancing temporal correlations among adjacent frames. Although numerous efforts have been made to optimize MCTF, MCTF still lacks flexibility in filtering for diverse video content, and its filtering efficiency is still limited. In this paper, we propose the Content-Aware MCTF method (CAMCTF) to enhance the filtering adaptability of MCTF for diverse video contents. The CAMCTF adaptively adjusts the filtering block sizes based on the Sum of Square Error (SSE) and Motion Vector (MV) information calculated during the motion estimation (ME) process in MCTF, and offset weighting method is applied to improve the prediction quality of filtering blocks. Performance was evaluated on top of Versatile Video Coding (VVC) reference software VTM-23.4. The VVC Common Test Conditions (CTC) [2] with QPs 22, 27, 32, 37 are used. As shown in Table 1, CAMCTF achieves an overall of 1.07% and 0.79% luma BD-rate gains in Random Access (RA) and Low Delay (LD) configurations, respectively. The encoder complexity is 112% and 110% for RA and LD configurations, respectively, without decoder complexity increasing.

Efficiently improving the accuracy of motion vector (MV) in merge candidate list is a critical issue in terms of the advanced inter coding technologies. The decoder side motion vector refinement (DMVR) and merge mode with motion vector differences (MMVD) are utilized to refine the MV obtained from merge mode in Versatile Video Coding (VVC). Nevertheless, both DMVR and MMVD operate under the assumption of symmetric motion when adjusting bi-prediction MV. This assumption may result in inaccuracies adjusting where the motion is asymmetric, leading to imprecise MV. To address this issue, we propose the asymmetric motion vector refinement (ASMVR) approach to refine asymmetric motion more accurately for future video coding in this paper. Specifically, ASMVR is formulated by the asymmetric MMVD (asy-MMVD) and asymmetric DMVR (asy-DMVR) schemes, which are compatible with MMVD and DMVR in VVC respectively. Four asymmetric MV refinement templates are devised to capture varying motion scenes, and the optimal one is derived through bilateral matching and rate-distortion optimization. Moreover, meticulously designed fast algorithms are implemented to bypass unnecessary candidate evaluations, thereby effectively reducing both encoding and decoding complexities. The simulation result shows that on top of the VVC Test Model (VTM-22.1), ASMVR achieves 1.52% BD-rate saving for random access (RA).

Congestion control has been a fundamental research focus in web transmission for over 30 years. However, with diverse network scenarios like cellular networks and WiFi, traditional models might no longer accurately describe current network conditions -- we empirically observe that the minimum round-trip time (RTTmin) still varies under different network conditions, challenging the assumption of its constancy in traditional models. In this paper, we model it as a normal distribution based on our measurements and propose a novel congestion control algorithm LingBo. LingBo consists of two phases: an offline trained decision model to achieve goals under different RTTmin distributions, and an online perception scheme to detect the current RTTmin distribution. We evaluate LingBo in various network environments and find it consistently performs well in terms of power metric and throughput compared to recent state-of-the-art baselines. Our code is available at https://github.com/thumedia/LingBo.

The leading learning-based rate control method, i.e., QARC, achieves state-of-the-art performances but fails to interpret the fundamental principles, and thus lacks the abilities to further improve itself efficiently. In this paper, we propose EQARC (Explainable QARC) via reconstructing QARC's modules, aiming to demystify how QARC works. In details, we first utilize a novel hybrid attention-based CNN+GRU model to re-characterize the original quality prediction network and reasonably replace the QARC's 1D-CNN layers with 2D-CNN layers. Using trace-driven experiment, we demonstrate the superiority of EQARC over existing state-of-the-art approaches. Next, we collect several useful information from each interpretable modules and learn the insight of EQARC. Following this step, we further propose AQARC (Advanced QARC), which is the light-weighted version of QARC. Experimental results show that AQARC achieves the same performances as the QARC with an overhead reduction of 90%. In short, through learning from deep learning, we generalize a rate control method which can both reach high performance and reduce computation cost.