Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper.

视频超分辨率（VSR）是从一系列低分辨率输入序列恢复高分辨率帧的任务。与单图超分辨率不同，VSR可以利用框架的时间信息来重建结果，并提供更多详细信息。最近，随着卷积神经网络（CNN）的快速发展，VSR任务引起了人们的关注，许多基于CNN的方法取得了显着的结果。但是，由于计算资源和运行时限制，只能将一些VSR方法应用于现实世界移动设备。在本文中，我们提出了一个\ textIt {基于滑动窗口的重复网络}（SWRN），该网络可以实时推断，同时仍能达到卓越的性能。具体而言，我们注意到视频帧应该具有可以帮助恢复细节的空间和时间关系，而关键点是如何提取和汇总信息。解决它，我们输入了三个相邻的帧，并利用隐藏状态来重复存储和更新重要的时间信息。我们在REDS数据集上的实验表明，所提出的方法可以很好地适应移动设备并产生视觉上令人愉悦的结果。

突发超级分辨率（SR）提供了从低质量图像恢复丰富细节的可能性。然而，由于实际应用中的低分辨率（LR）图像具有多种复杂和未知的降级，所以现有的非盲（例如，双臂）设计的网络通常导致恢复高分辨率（HR）图像的严重性能下降。此外，处理多重未对准的嘈杂的原始输入也是具有挑战性的。在本文中，我们解决了从现代手持设备获取的原始突发序列重建HR图像的问题。中央观点是一个内核引导策略，可以用两个步骤解决突发SR：内核建模和HR恢复。前者估计来自原始输入的突发内核，而后者基于估计的内核预测超分辨图像。此外，我们引入了内核感知可变形对准模块，其可以通过考虑模糊的前沿而有效地对准原始图像。对综合和现实世界数据集的广泛实验表明，所提出的方法可以在爆发SR问题中对最先进的性能进行。

Deploying reliable deep learning techniques in interdisciplinary applications needs learned models to output accurate and ({even more importantly}) explainable predictions. Existing approaches typically explicate network outputs in a post-hoc fashion, under an implicit assumption that faithful explanations come from accurate predictions/classifications. We have an opposite claim that explanations boost (or even determine) classification. That is, end-to-end learning of explanation factors to augment discriminative representation extraction could be a more intuitive strategy to inversely assure fine-grained explainability, e.g., in those neuroimaging and neuroscience studies with high-dimensional data containing noisy, redundant, and task-irrelevant information. In this paper, we propose such an explainable geometric deep network dubbed as NeuroExplainer, with applications to uncover altered infant cortical development patterns associated with preterm birth. Given fundamental cortical attributes as network input, our NeuroExplainer adopts a hierarchical attention-decoding framework to learn fine-grained attentions and respective discriminative representations to accurately recognize preterm infants from term-born infants at term-equivalent age. NeuroExplainer learns the hierarchical attention-decoding modules under subject-level weak supervision coupled with targeted regularizers deduced from domain knowledge regarding brain development. These prior-guided constraints implicitly maximizes the explainability metrics (i.e., fidelity, sparsity, and stability) in network training, driving the learned network to output detailed explanations and accurate classifications. Experimental results on the public dHCP benchmark suggest that NeuroExplainer led to quantitatively reliable explanation results that are qualitatively consistent with representative neuroimaging studies.

In the field of cross-modal retrieval, single encoder models tend to perform better than dual encoder models, but they suffer from high latency and low throughput. In this paper, we present a dual encoder model called BagFormer that utilizes a cross modal interaction mechanism to improve recall performance without sacrificing latency and throughput. BagFormer achieves this through the use of bag-wise interactions, which allow for the transformation of text to a more appropriate granularity and the incorporation of entity knowledge into the model. Our experiments demonstrate that BagFormer is able to achieve results comparable to state-of-the-art single encoder models in cross-modal retrieval tasks, while also offering efficient training and inference with 20.72 times lower latency and 25.74 times higher throughput.

Body Mass Index (BMI), age, height and weight are important indicators of human health conditions, which can provide useful information for plenty of practical purposes, such as health care, monitoring and re-identification. Most existing methods of health indicator prediction mainly use front-view body or face images. These inputs are hard to be obtained in daily life and often lead to the lack of robustness for the models, considering their strict requirements on view and pose. In this paper, we propose to employ gait videos to predict health indicators, which are more prevalent in surveillance and home monitoring scenarios. However, the study of health indicator prediction from gait videos using deep learning was hindered due to the small amount of open-sourced data. To address this issue, we analyse the similarity and relationship between pose estimation and health indicator prediction tasks, and then propose a paradigm enabling deep learning for small health indicator datasets by pre-training on the pose estimation task. Furthermore, to better suit the health indicator prediction task, we bring forward Global-Local Aware aNd Centrosymmetric Encoder (GLANCE) module. It first extracts local and global features by progressive convolutions and then fuses multi-level features by a centrosymmetric double-path hourglass structure in two different ways. Experiments demonstrate that the proposed paradigm achieves state-of-the-art results for predicting health indicators on MoVi, and that the GLANCE module is also beneficial for pose estimation on 3DPW.

We present IMAS, a method that segments the primary objects in videos without manual annotation in training or inference. Previous methods in unsupervised video object segmentation (UVOS) have demonstrated the effectiveness of motion as either input or supervision for segmentation. However, motion signals may be uninformative or even misleading in cases such as deformable objects and objects with reflections, causing unsatisfactory segmentation. In contrast, IMAS achieves Improved UVOS with Motion-Appearance Synergy. Our method has two training stages: 1) a motion-supervised object discovery stage that deals with motion-appearance conflicts through a learnable residual pathway; 2) a refinement stage with both low- and high-level appearance supervision to correct model misconceptions learned from misleading motion cues. Additionally, we propose motion-semantic alignment as a model-agnostic annotation-free hyperparam tuning method. We demonstrate its effectiveness in tuning critical hyperparams previously tuned with human annotation or hand-crafted hyperparam-specific metrics. IMAS greatly improves the segmentation quality on several common UVOS benchmarks. For example, we surpass previous methods by 8.3% on DAVIS16 benchmark with only standard ResNet and convolutional heads. We intend to release our code for future research and applications.

Current audio-visual separation methods share a standard architecture design where an audio encoder-decoder network is fused with visual encoding features at the encoder bottleneck. This design confounds the learning of multi-modal feature encoding with robust sound decoding for audio separation. To generalize to a new instrument: one must finetune the entire visual and audio network for all musical instruments. We re-formulate visual-sound separation task and propose Instrument as Query (iQuery) with a flexible query expansion mechanism. Our approach ensures cross-modal consistency and cross-instrument disentanglement. We utilize "visually named" queries to initiate the learning of audio queries and use cross-modal attention to remove potential sound source interference at the estimated waveforms. To generalize to a new instrument or event class, drawing inspiration from the text-prompt design, we insert an additional query as an audio prompt while freezing the attention mechanism. Experimental results on three benchmarks demonstrate that our iQuery improves audio-visual sound source separation performance.

Learning generalizable insertion skills in a data-efficient manner has long been a challenge in the robot learning community. While the current state-of-the-art methods with reinforcement learning (RL) show promising performance in acquiring manipulation skills, the algorithms are data-hungry and hard to generalize. To overcome the issues, in this paper we present Prim-LAfD, a simple yet effective framework to learn and adapt primitive-based insertion skills from demonstrations. Prim-LAfD utilizes black-box function optimization to learn and adapt the primitive parameters leveraging prior experiences. Human demonstrations are modeled as dense rewards guiding parameter learning. We validate the effectiveness of the proposed method on eight peg-hole and connector-socket insertion tasks. The experimental results show that our proposed framework takes less than one hour to acquire the insertion skills and as few as fifteen minutes to adapt to an unseen insertion task on a physical robot.

Partial label learning (PLL) is a typical weakly supervised learning, where each sample is associated with a set of candidate labels. The basic assumption of PLL is that the ground-truth label must reside in the candidate set. However, this assumption may not be satisfied due to the unprofessional judgment of the annotators, thus limiting the practical application of PLL. In this paper, we relax this assumption and focus on a more general problem, noisy PLL, where the ground-truth label may not exist in the candidate set. To address this challenging problem, we further propose a novel framework called "Automatic Refinement Network (ARNet)". Our method consists of multiple rounds. In each round, we purify the noisy samples through two key modules, i.e., noisy sample detection and label correction. To guarantee the performance of these modules, we start with warm-up training and automatically select the appropriate correction epoch. Meanwhile, we exploit data augmentation to further reduce prediction errors in ARNet. Through theoretical analysis, we prove that our method is able to reduce the noise level of the dataset and eventually approximate the Bayes optimal classifier. To verify the effectiveness of ARNet, we conduct experiments on multiple benchmark datasets. Experimental results demonstrate that our ARNet is superior to existing state-of-the-art approaches in noisy PLL. Our code will be made public soon.