End-to-end autonomous driving provides a feasible way to automatically maximize overall driving system performance by directly mapping the raw pixels from a front-facing camera to control signals. Recent advanced methods construct a latent world model to map the high dimensional observations into compact latent space. However, the latent states embedded by the world model proposed in previous works may contain a large amount of task-irrelevant information, resulting in low sampling efficiency and poor robustness to input perturbations. Meanwhile, the training data distribution is usually unbalanced, and the learned policy is hard to cope with the corner cases during the driving process. To solve the above challenges, we present a semantic masked recurrent world model (SEM2), which introduces a latent filter to extract key task-relevant features and reconstruct a semantic mask via the filtered features, and is trained with a multi-source data sampler, which aggregates common data and multiple corner case data in a single batch, to balance the data distribution. Extensive experiments on CARLA show that our method outperforms the state-of-the-art approaches in terms of sample efficiency and robustness to input permutations.
translated by 谷歌翻译
Weakly-supervised temporal action localization (WTAL) learns to detect and classify action instances with only category labels. Most methods widely adopt the off-the-shelf Classification-Based Pre-training (CBP) to generate video features for action localization. However, the different optimization objectives between classification and localization, make temporally localized results suffer from the serious incomplete issue. To tackle this issue without additional annotations, this paper considers to distill free action knowledge from Vision-Language Pre-training (VLP), since we surprisingly observe that the localization results of vanilla VLP have an over-complete issue, which is just complementary to the CBP results. To fuse such complementarity, we propose a novel distillation-collaboration framework with two branches acting as CBP and VLP respectively. The framework is optimized through a dual-branch alternate training strategy. Specifically, during the B step, we distill the confident background pseudo-labels from the CBP branch; while during the F step, the confident foreground pseudo-labels are distilled from the VLP branch. And as a result, the dual-branch complementarity is effectively fused to promote a strong alliance. Extensive experiments and ablation studies on THUMOS14 and ActivityNet1.2 reveal that our method significantly outperforms state-of-the-art methods.
translated by 谷歌翻译
The statistical heterogeneity of the non-independent and identically distributed (non-IID) data in local clients significantly limits the performance of federated learning. Previous attempts like FedProx, SCAFFOLD, MOON, FedNova and FedDyn resort to an optimization perspective, which requires an auxiliary term or re-weights local updates to calibrate the learning bias or the objective inconsistency. However, in addition to previous explorations for improvement in federated averaging, our analysis shows that another critical bottleneck is the poorer optima of client models in more heterogeneous conditions. We thus introduce a data-driven approach called FedSkip to improve the client optima by periodically skipping federated averaging and scattering local models to the cross devices. We provide theoretical analysis of the possible benefit from FedSkip and conduct extensive experiments on a range of datasets to demonstrate that FedSkip achieves much higher accuracy, better aggregation efficiency and competing communication efficiency. Source code is available at: https://github.com/MediaBrain-SJTU/FedSkip.
translated by 谷歌翻译
低光视频增强(LLVE)是许多应用程序,例如拍摄和自动驾驶,是一项重要但艰巨的任务。与单图像低光增强不同,大多数LLVE方法都利用相邻帧的时间信息来恢复颜色并删除目标框架的噪声。但是,这些算法基于多帧对齐和增强的框架,在遇到极端低光或快速运动时可能会产生多帧融合工件。在本文中,受到低潜伏期和高动态事件范围的启发,我们使用来自多个帧的合成事件来指导低光视频的增强和恢复。我们的方法包含三个阶段:1)事件合成和增强,2)事件和图像融合,以及3)低光增强。在此框架中,我们分别为第二阶段和第三阶段设计了两个新型模块(事件图像融合变换和事件引导的双分支)。广泛的实验表明,我们的方法在合成数据集和真实LLVE数据集上都优于现有的低光视频或单个图像增强方法。
translated by 谷歌翻译
在多模式的多代理轨迹预测中,尚未完全解决两个主要挑战:1)如何测量相互作用模块引起的不确定性,从而导致多个试剂的预测轨迹之间引起相关性; 2)如何对多个预测进行排名并选择最佳预测轨迹。为了应对这些挑战,这项工作首先提出了一个新颖的概念,协作不确定性(CU),该概念模拟了互动模块引起的不确定性。然后,我们使用原始置换量等不确定性估计器来构建一般的CU感知回归框架,以完成回归和不确定性估计任务。此外,我们将提出的框架应用于当前的SOTA多代理多模式预测系统作为插件模块,该模块使SOTA系统能够达到1)估计多代理多模式轨迹预测任务的不确定性; 2)对多个预测进行排名,并根据估计的不确定性选择最佳预测。我们对合成数据集和两个公共大规模多代理轨迹预测基准进行了广泛的实验。实验表明:1)在合成数据集上,Cu-Aware回归框架允许模型适当地近似地面真相拉普拉斯分布; 2)在多代理轨迹预测基准上,Cu-Aware回归框架稳步帮助SOTA系统改善了其性能。特别是,提出的框架帮助Vectornet在Nuscenes数据集中所选最佳预测的最终位移误差方面提高了262 cm; 3)对于多机构多模式轨迹预测系统,预测不确定性与未来随机性呈正相关; 4)估计的CU值与代理之间的交互式信息高度相关。
translated by 谷歌翻译
多模式情绪识别的研究和应用最近变得越来越流行。但是,多模式情绪识别面临缺乏数据的挑战。为了解决这个问题,我们建议使用转移学习,哪些人利用最先进的预培训模型,包括WAV2VEC 2.0和BERT来执行此任务。探索了多级融合方法,包括基于共发的早期融合和与在两个嵌入训练的模型的后期融合。此外,还提出了一个多范围的框架,它不仅提取了帧级的语音嵌入,还提出了细分级别的嵌入,包括电话,音节和文字级语音嵌入,以进一步提高性能。通过将基于同时的早期融合模型和晚期融合模型与多粒性特征提取框架相结合,我们获得的结果使IEMOCAP数据集上的最佳基线方法优于最佳基线方法未加权准确性(UA)。
translated by 谷歌翻译
本文考虑了快速MRI重建的问题。我们提出了一个基于变压器的新型框架,用于直接处理K空间中稀疏采样的信号,超出了像Convnets一样的常规网格的限制。我们采用频谱图的隐式表示,将空间坐标视为输入,并动态查询部分观察到的测量值以完成频谱图,即学习K空间中的电感偏置。为了在计算成本和重建质量之间保持平衡,我们分别建立了一个具有低分辨率和高分辨率解码器的层次结构。为了验证我们提出的模块的必要性,我们在两个公共数据集上进行了广泛的实验,并表现出优于最先进方法的卓越或可比性。
translated by 谷歌翻译
自我监督的学习在表示视觉和文本数据的表示方面取得了巨大的成功。但是,当前的方法主要在经过良好策划的数据集中验证,这些数据集未显示现实世界的长尾分布。在损失的角度或模型观点中,重新平衡的重新平衡是为了考虑自我监督的长尾学习的最新尝试,类似于被监督的长尾学习中的范式。然而,没有标签的帮助,由于尾巴样品发现或启发式结构设计的限制,这些探索并未显示出预期的明显希望。与以前的作品不同,我们从替代角度(即数据角度)探索了这个方向,并提出了一种新颖的增强对比度学习(BCL)方法。具体而言,BCL利用深神经网络的记忆效果自动推动对比度学习中样本视图的信息差异,这更有效地增强了标签 - unaware环境中的长尾学习。对一系列基准数据集进行的广泛实验证明了BCL对几种最新方法的有效性。我们的代码可在https://github.com/mediabrain-sjtu/bcl上找到。
translated by 谷歌翻译
我们展示了MVLayoutNet,是来自多视图全景的整体三维重建端到端网络。我们的核心贡献是无缝地将学习的单目布局估计和多视图立体声(MV)结合起来,以便在3D和图像空间中准确地重建。我们共同列出布局模块以产生初始布局和新型MVS模块,以获得精确的布局几何形状。与标准MVSNET [33]不同,我们的MVS模块采用新建的布局成本卷,其在相同的深度层中聚合到相应的布局元件中的多视图成本。我们还提供了一种基于注意的方案,指导MVS模块专注于结构区域。这种设计考虑了本地像素级成本和全球整体信息,以便更好地重建。实验表明,我们的方法在2D-3D-S [1]和Zind [5]数据集中,在深度RMSE方面以21.7%和20.6%表示最先进的。最后,我们的方法导致连贯的布局几何,使整个场景的重建能够。
translated by 谷歌翻译
手写的数学表达式识别旨在自动生成来自给定图像的乳胶序列。目前,基于注意的编码器 - 解码器模型被广泛用于此任务。它们通常以左右(L2R)方式生成目标序列,留下左右(R2L)上下文未分发。在本文中,我们提出了一种基于聚合的双向互访网络(ABM),其包括一个共享编码器和两个并行逆解码器(L2R和R2L)组成。通过相互蒸馏增强了两个解码器,其涉及每个训练步骤的一对一知识转移,从而充分利用来自两个反向的互补信息。此外,为了处理各种规模的数学符号,提出了注意聚合模块(AAM)以有效地集成了多尺度覆盖关注。值得注意的是,在推理阶段,考虑到模型已经从两个反向方向学习知识,我们只使用L2R分支推断,保持原始参数大小和推断速度。广泛的实验表明,我们的拟议方法在2016年克罗欧2014年达到56.85%的识别准确性,52.92%,在克罗欧2019年的53.96%,没有数据增强和模型集合,大大优于最先进的方法。源代码可在补充材料中获得。
translated by 谷歌翻译