Model efficiency has become increasingly important in computer vision. In this paper, we systematically study neural network architecture design choices for object detection and propose several key optimizations to improve efficiency. First, we propose a weighted bi-directional feature pyramid network (BiFPN), which allows easy and fast multiscale feature fusion; Second, we propose a compound scaling method that uniformly scales the resolution, depth, and width for all backbone, feature network, and box/class prediction networks at the same time. Based on these optimizations and better backbones, we have developed a new family of object detectors, called EfficientDet, which consistently achieve much better efficiency than prior art across a wide spectrum of resource constraints. In particular, with singlemodel and single-scale, our EfficientDet-D7 achieves stateof-the-art 55.1 AP on COCO test-dev with 77M parameters and 410B FLOPs 1 , being 4x -9x smaller and using 13x -42x fewer FLOPs than previous detectors. Code is available at https://github.com/google/automl/tree/ master/efficientdet.
translated by 谷歌翻译
人们普遍认为,对于准确的语义细分,必须使用昂贵的操作(例如,非常卷积)结合使用昂贵的操作(例如非常卷积),从而导致缓慢的速度和大量的内存使用。在本文中,我们质疑这种信念,并证明既不需要高度的内部决议也不是必需的卷积。我们的直觉是,尽管分割是一个每像素的密集预测任务,但每个像素的语义通常都取决于附近的邻居和遥远的环境。因此,更强大的多尺度功能融合网络起着至关重要的作用。在此直觉之后,我们重新访问常规的多尺度特征空间(通常限制为P5),并将其扩展到更丰富的空间,最小的P9,其中最小的功能仅为输入大小的1/512,因此具有很大的功能接受场。为了处理如此丰富的功能空间,我们利用最近的BIFPN融合了多尺度功能。基于这些见解,我们开发了一个简化的分割模型,称为ESEG,该模型既没有内部分辨率高,也没有昂贵的严重卷积。也许令人惊讶的是,与多个数据集相比,我们的简单方法可以以比以前的艺术更快地实现更高的准确性。在实时设置中,ESEG-Lite-S在189 fps的CityScapes [12]上达到76.0%MIOU,表现优于更快的[9](73.1%MIOU时为170 fps)。我们的ESEG-LITE-L以79 fps的速度运行,达到80.1%MIOU,在很大程度上缩小了实时和高性能分割模型之间的差距。
translated by 谷歌翻译
在传统的对象检测框架中,从图像识别模型继承的骨干体提取了深层特征,然后颈部模块融合了这些潜在特征,以在不同的尺度上捕获信息。由于对象检测的分辨率比图像识别大得多,因此骨干的计算成本通常主导了总推断成本。这种沉重的背部设计范式主要是由于历史遗产将图像识别模型传输到对象检测时,而不是端到端的优化设计以进行对象检测。在这项工作中,我们表明这种范式确实导致了亚最佳对象检测模型。为此,我们提出了一种新型的重颈范式,长颈鹿,这是一个类似长颈鹿的网络,用于有效的对象检测。长颈鹿使用极轻的骨干和非常深的颈部模块,可同时同时在不同的空间尺度以及不同级别的潜在语义之间进行密集的信息交换。该设计范式允许检测器即使在网络的早期阶段,也可以在相同的优先级处理高级语义信息和低级空间信息,从而使其在检测任务中更有效。对多个流行对象检测基准的数值评估表明,长颈鹿在广泛的资源约束中始终优于先前的SOTA模型。源代码可在https://github.com/jyqi/giraffedet上获得。
translated by 谷歌翻译
Current state-of-the-art convolutional architectures for object detection are manually designed. Here we aim to learn a better architecture of feature pyramid network for object detection. We adopt Neural Architecture Search and discover a new feature pyramid architecture in a novel scalable search space covering all cross-scale connections. The discovered architecture, named NAS-FPN, consists of a combination of top-down and bottom-up connections to fuse features across scales. NAS-FPN, combined with various backbone models in the RetinaNet framework, achieves better accuracy and latency tradeoff compared to state-ofthe-art object detection models. NAS-FPN improves mobile detection accuracy by 2 AP compared to state-of-the-art SS-DLite with MobileNetV2 model in [32] and achieves 48.3 AP which surpasses Mask R-CNN [10] detection accuracy with less computation time.
translated by 谷歌翻译
In this report, we present a fast and accurate object detection method dubbed DAMO-YOLO, which achieves higher performance than the state-of-the-art YOLO series. DAMO-YOLO is extended from YOLO with some new technologies, including Neural Architecture Search (NAS), efficient Reparameterized Generalized-FPN (RepGFPN), a lightweight head with AlignedOTA label assignment, and distillation enhancement. In particular, we use MAE-NAS, a method guided by the principle of maximum entropy, to search our detection backbone under the constraints of low latency and high performance, producing ResNet-like / CSP-like structures with spatial pyramid pooling and focus modules. In the design of necks and heads, we follow the rule of "large neck, small head". We import Generalized-FPN with accelerated queen-fusion to build the detector neck and upgrade its CSPNet with efficient layer aggregation networks (ELAN) and reparameterization. Then we investigate how detector head size affects detection performance and find that a heavy neck with only one task projection layer would yield better results. In addition, AlignedOTA is proposed to solve the misalignment problem in label assignment. And a distillation schema is introduced to improve performance to a higher level. Based on these new techs, we build a suite of models at various scales to meet the needs of different scenarios, i.e., DAMO-YOLO-Tiny/Small/Medium. They can achieve 43.0/46.8/50.0 mAPs on COCO with the latency of 2.78/3.83/5.62 ms on T4 GPUs respectively. The code is available at https://github.com/tinyvision/damo-yolo.
translated by 谷歌翻译
Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient. We demonstrate the effectiveness of this method on scaling up MobileNets and ResNet.To go even further, we use neural architecture search to design a new baseline network and scale it up to obtain a family of models, called EfficientNets, which achieve much better accuracy and efficiency than previous ConvNets. In particular, our EfficientNet-B7 achieves state-of-the-art 84.3% top-1 accuracy on ImageNet, while being 8.4x smaller and 6.1x faster on inference than the best existing ConvNet. Our EfficientNets also transfer well and achieve state-of-the-art accuracy on CIFAR-100 (91.7%), Flowers (98.8%), and 3 other transfer learning datasets, with an order of magnitude fewer parameters. Source code is at https: //github.com/tensorflow/tpu/tree/ master/models/official/efficientnet.
translated by 谷歌翻译
Adder神经网络(Addernets)在图像分类上表现出令人印象深刻的性能,只有加法操作,比使用乘法建立的传统卷积神经网络更节能。与分类相比,对通过Addernets降低现代对象探测器的能耗的强烈需求,例如自主驾驶和面部检测。在本文中,我们提出了对物体检测的addernets的实证研究。我们首先揭示了预先训练的加法器骨架中的批量归一化统计,不应冻结,因为Addernets的相对较大的特征方差。此外,我们在颈部中插入更多的快捷方式连接,并设计一个新的特征融合架构,以避免加法器层的稀疏功能。我们展示了广泛的消融研究,探讨了加法器探测器的几种设计选择。与最先进的比较在Coco和Pascal VOC基准上进行。具体而言,所提出的加法器FCOS在Coco Val集上实现了37.8 \%AP,展示了卷积对应物的相当性能,具有约1.4倍的能量减少。
translated by 谷歌翻译
更好的准确性和效率权衡在对象检测中是一个具有挑战性的问题。在这项工作中,我们致力于研究对象检测的关键优化和神经网络架构选择,以提高准确性和效率。我们调查了无锚策略对轻质对象检测模型的适用性。我们增强了骨干结构并设计了颈部的轻质结构,从而提高了网络的特征提取能力。我们改善标签分配策略和损失功能,使培训更稳定和高效。通过这些优化,我们创建了一个名为PP-Picodet的新的实时对象探测器系列,这在移动设备的对象检测上实现了卓越的性能。与其他流行型号相比,我们的模型在准确性和延迟之间实现了更好的权衡。 Picodet-s只有0.99m的参数达到30.6%的地图,它是地图的绝对4.8%,同时与yolox-nano相比将移动CPU推理延迟减少55%,并且与Nanodet相比,MAP的绝对改善了7.1%。当输入大小为320时,它在移动臂CPU上达到123个FPS(使用桨Lite)。Picodet-L只有3.3M参数,达到40.9%的地图,这是地图的绝对3.7%,比yolov5s更快44% 。如图1所示,我们的模型远远优于轻量级对象检测的最先进的结果。代码和预先训练的型号可在https://github.com/paddlepaddle/paddledentions提供。
translated by 谷歌翻译
由于卷积在提取物体的局部上下文中,在过去十年中,对象检测在过去十年中取得了重大进展。但是,对象的尺度是多样的,当前卷积只能处理单尺度输入。因此,传统卷积具有固定接收场在处理这种规模差异问题方面的能力受到限制。多尺度功能表示已被证明是缓解规模差异问题的有效方法。最近的研究主要与某些量表或各个尺度的总体特征采用部分联系,并专注于整个量表的全球信息。但是,跨空间和深度维度的信息被忽略了。受此启发,我们提出了多尺度卷积(MSCONV)来解决此问题。同时考虑到量表,空间和深度信息,MSCONV能够更全面地处理多尺度输入。 MSCONV是有效的,并且在计算上是有效的,只有少量计算成本增加。对于大多数单阶段对象探测器,在检测头中用MSCONV代替传统的卷积可以带来AP的2.5 \%改进(在Coco 2017数据集上),只有3 \%的拖鞋增加了。 MSCONV对于两阶段对象探测器也具有灵活性和有效性。当扩展到主流两阶段对象检测器时,MSCONV的AP可以提高3.0 \%。我们在单尺度测试下的最佳模型在Coco 2017上实现了48.9 \%AP,\ textit {test-dev} Split,它超过了许多最新方法。
translated by 谷歌翻译
本文提出了平行残留的双融合特征金字塔网络(PRB-FPN),以快速准确地单光对象检测。特征金字塔(FP)在最近的视觉检测中被广泛使用,但是由于汇总转换,FP的自上而下的途径无法保留准确的定位。随着使用更多层的更深骨干,FP的优势被削弱了。此外,它不能同时准确地检测到小物体。为了解决这些问题,我们提出了一种新的并行FP结构,具有双向(自上而下和自下而上)的融合以及相关的改进,以保留高质量的特征以进行准确定位。我们提供以下设计改进:(1)具有自下而上的融合模块(BFM)的平行分歧FP结构,以高精度立即检测小物体和大对象。 (2)串联和重组(CORE)模块为特征融合提供了自下而上的途径,该途径导致双向融合FP,可以从低层特征图中恢复丢失的信息。 (3)进一步纯化核心功能以保留更丰富的上下文信息。自上而下和自下而上的途径中的这种核心净化只能在几次迭代中完成。 (4)将残留设计添加到核心中,导致了一个新的重核模块,该模块可以轻松训练和集成,并具有更深入或更轻的骨架。所提出的网络可在UAVDT17和MS COCO数据集上实现最新性能。代码可在https://github.com/pingyang1117/prbnet_pytorch上找到。
translated by 谷歌翻译
大多数现有的深神经网络都是静态的,这意味着它们只能以固定的复杂性推断。但资源预算可以大幅度不同。即使在一个设备上,实惠预算也可以用不同的场景改变,并且对每个所需预算的反复培训网络是非常昂贵的。因此,在这项工作中,我们提出了一种称为Mutualnet的一般方法,以训练可以以各种资源约束运行的单个网络。我们的方法列举了具有各种网络宽度和输入分辨率的模型配置队列。这种相互学习方案不仅允许模型以不同的宽度分辨率配置运行,而且还可以在这些配置之间传输独特的知识,帮助模型来学习更强大的表示。 Mutualnet是一般的培训方法,可以应用于各种网络结构(例如,2D网络:MobileNets,Reset,3D网络:速度,X3D)和各种任务(例如,图像分类,对象检测,分段和动作识别),并证明了实现各种数据集的一致性改进。由于我们只培训了这一模型,它对独立培训多种型号而言,它也大大降低了培训成本。令人惊讶的是,如果动态资源约束不是一个问题,则可以使用Mutualnet来显着提高单个网络的性能。总之,Mutualnet是静态和自适应,2D和3D网络的统一方法。代码和预先训练的模型可用于\ url {https://github.com/tayang1122/mutualnet}。
translated by 谷歌翻译
本文分析了面部检测体系结构的设计选择,以提高计算成本和准确性之间的效率。具体而言,我们重新检查了标准卷积块作为面部检测的轻质骨干结构的有效性。与当前的轻质体系结构设计的趋势(大量利用了可分开的卷积层)不同,我们表明,使用类似的参数大小时,大量通道绕的标准卷积层可以实现更好的准确性和推理速度。关于目标数据域的特征的分析,该观察结果得到了支持。根据我们的观察,我们建议使用高度降低的通道使用Resnet,与其他移动友好网络(例如Mobilenet-V1,-V2,-V3)相比,它具有高度效率。从广泛的实验中,我们表明所提出的主链可以以更快的推理速度替换最先进的面部检测器的主链。此外,我们进一步提出了一种最大化检测性能的新功能聚合方法。我们提出的检测器ERESFD获得了更宽的面部硬子子集的80.4%地图,该图仅需37.7 ms即可在CPU上进行VGA图像推断。代码将在https://github.com/clovaai/eresfd上找到。
translated by 谷歌翻译
在对象检测模型中,检测骨干机消耗超过一半的整体推理成本。最近的研究试图通过在神经结构搜索(NAS)的帮助下优化骨干架构来降低这一成本。然而,对象检测的现有NAS方法需要数百至数千个GPU小时的搜索,使它们在快节奏的研究和开发中不切实际。在这项工作中,我们提出了一种新的零射NAS方法来解决这个问题。所提出的方法,命名为Zendet,在不训练网络参数的情况下自动设计有效的检测骨干网,从而降低了架构设计成本,几乎归零但提供了最先进的(SOTA)性能。在引擎盖下,Zendet最大化了检测骨干的差分熵,导致对象检测的更好的特征提取器,在相同的计算预算下。在仅为全自动设计的一个GPU日之后,Zendet在多个检测基准数据集上创新了SOTA检测骨干,具有很少的人为干预。与Reset-50个骨干相比,Zendet在Map中使用相同数量的拖波/参数时更好地+ 2.0%,并且在同一地图上的NVIDIA V100速度快1.54倍。稍后将发布代码和预先训练的型号。
translated by 谷歌翻译
为了实现不断增长的准确性,通常会开发大型和复杂的神经网络。这样的模型需要高度的计算资源,因此不能在边缘设备上部署。由于它们在几个应用领域的有用性,建立资源有效的通用网络非常感兴趣。在这项工作中,我们努力有效地结合了CNN和变压器模型的优势,并提出了一种新的有效混合体系结构。特别是在EDGENEXT中,我们引入了分裂深度转置注意力(SDTA)编码器,该编码器将输入张量分解为多个通道组,并利用深度旋转以及跨通道维度的自我注意力,以隐含地增加接受场并编码多尺度特征。我们在分类,检测和分割任务上进行的广泛实验揭示了所提出的方法的优点,优于相对较低的计算要求的最先进方法。我们具有130万参数的EDGENEXT模型在Imagenet-1k上达到71.2 \%TOP-1的精度,超过移动设备的绝对增益为2.2 \%,而拖鞋减少了28 \%。此外,我们具有560万参数的EDGENEXT模型在Imagenet-1k上达到了79.4 \%TOP-1的精度。代码和模型可在https://t.ly/_vu9上公开获得。
translated by 谷歌翻译
现代的高性能语义分割方法采用沉重的主链和扩张的卷积来提取相关特征。尽管使用上下文和语义信息提取功能对于分割任务至关重要,但它为实时应用程序带来了内存足迹和高计算成本。本文提出了一种新模型,以实现实时道路场景语义细分的准确性/速度之间的权衡。具体来说,我们提出了一个名为“比例吸引的条带引导特征金字塔网络”(s \ textsuperscript {2} -fpn)的轻巧模型。我们的网络由三个主要模块组成:注意金字塔融合(APF)模块,比例吸引条带注意模块(SSAM)和全局特征Upsample(GFU)模块。 APF采用了注意力机制来学习判别性多尺度特征,并有助于缩小不同级别之间的语义差距。 APF使用量表感知的关注来用垂直剥离操作编码全局上下文,并建模长期依赖性,这有助于将像素与类似的语义标签相关联。此外,APF还采用频道重新加权块(CRB)来强调频道功能。最后,S \ TextSuperScript {2} -fpn的解码器然后采用GFU,该GFU用于融合APF和编码器的功能。已经对两个具有挑战性的语义分割基准进行了广泛的实验,这表明我们的方法通过不同的模型设置实现了更好的准确性/速度权衡。提出的模型已在CityScapes Dataset上实现了76.2 \%miou/87.3fps,77.4 \%miou/67fps和77.8 \%miou/30.5fps,以及69.6 \%miou,71.0 miou,71.0 \%miou,和74.2 \%\%\%\%\%\%。 miou在Camvid数据集上。这项工作的代码将在\ url {https://github.com/mohamedac29/s2-fpn提供。
translated by 谷歌翻译
神经结构搜索(NAS)已被广泛采用设计准确,高效的图像分类模型。但是,将NAS应用于新的计算机愿景任务仍然需要大量的努力。这是因为1)以前的NAS研究已经过度优先考虑图像分类,同时在很大程度上忽略了其他任务; 2)许多NAS工作侧重于优化特定于任务特定的组件,这些组件不能有利地转移到其他任务; 3)现有的NAS方法通常被设计为“Proxyless”,需要大量努力与每个新任务的培训管道集成。为了解决这些挑战,我们提出了FBNetv5,这是一个NAS框架,可以在各种视觉任务中寻找神经架构,以降低计算成本和人力努力。具体而言,我们设计1)一个简单但包容性和可转换的搜索空间; 2)用目标任务培训管道解开的多址搜索过程; 3)一种算法,用于同时搜索具有计算成本不可知的多个任务的架构到任务数。我们评估所提出的FBNetv5目标三个基本视觉任务 - 图像分类,对象检测和语义分割。 FBNETV5在单一搜索中搜索的模型在所有三个任务中都表现优于先前的议定书 - 现有技术:图像分类(例如,与FBNetv3相比,在与FBNetv3相比的同一拖鞋下的1 + 1.3%Imageet Top-1精度。 (例如,+ 1.8%较高的Ade20k Val。Miou比SegFormer为3.6倍的拖鞋),对象检测(例如,+ 1.1%Coco Val。与yolox相比,拖鞋的1.2倍的地图。
translated by 谷歌翻译
锥体网络是多尺度对象检测的标准方法。当前对特征金字塔网络的研究通常采用层连接来从特征层次结构的某些级别收集特征,并且不考虑它们之间的显着差异。我们提出了一个更好的特征金字塔网络的体系结构,称为选择性多尺度学习(SMSL),以解决此问题。SMSL高效且泛滥,可以将其集成到单阶段和两阶段检测器中以提高检测性能,几乎没有额外的推理成本。视网膜与SMSL的结合获得了可可数据集的AP(从39.1 \%到40.9 \%)的1.8 \%改进。与SMSL集成时,两阶段探测器的AP可以提高1.0 \%。
translated by 谷歌翻译
Designing convolutional neural networks (CNN) for mobile devices is challenging because mobile models need to be small and fast, yet still accurate. Although significant efforts have been dedicated to design and improve mobile CNNs on all dimensions, it is very difficult to manually balance these trade-offs when there are so many architectural possibilities to consider. In this paper, we propose an automated mobile neural architecture search (MNAS) approach, which explicitly incorporate model latency into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency. Unlike previous work, where latency is considered via another, often inaccurate proxy (e.g., FLOPS), our approach directly measures real-world inference latency by executing the model on mobile phones. To further strike the right balance between flexibility and search space size, we propose a novel factorized hierarchical search space that encourages layer diversity throughout the network. Experimental results show that our approach consistently outperforms state-of-the-art mobile CNN models across multiple vision tasks. On the ImageNet classification task, our MnasNet achieves 75.2% top-1 accuracy with 78ms latency on a Pixel phone, which is 1.8× faster than MobileNetV2 [29] with 0.5% higher accuracy and 2.3× faster than NASNet [36] with 1.2% higher accuracy. Our MnasNet also achieves better mAP quality than MobileNets for COCO object detection. Code is at https://github.com/tensorflow/tpu/ tree/master/models/official/mnasnet.
translated by 谷歌翻译
Building instance segmentation models that are dataefficient and can handle rare object categories is an important challenge in computer vision. Leveraging data augmentations is a promising direction towards addressing this challenge. Here, we perform a systematic study of the Copy-Paste augmentation (e.g., [13,12]) for instance segmentation where we randomly paste objects onto an image. Prior studies on Copy-Paste relied on modeling the surrounding visual context for pasting the objects. However, we find that the simple mechanism of pasting objects randomly is good enough and can provide solid gains on top of strong baselines. Furthermore, we show Copy-Paste is additive with semi-supervised methods that leverage extra data through pseudo labeling (e.g. self-training). On COCO instance segmentation, we achieve 49.1 mask AP and 57.3 box AP, an improvement of +0.6 mask AP and +1.5 box AP over the previous state-of-the-art. We further demonstrate that Copy-Paste can lead to significant improvements on the LVIS benchmark. Our baseline model outperforms the LVIS 2020 Challenge winning entry by +3.6 mask AP on rare categories.
translated by 谷歌翻译
This paper introduces EfficientNetV2, a new family of convolutional networks that have faster training speed and better parameter efficiency than previous models. To develop these models, we use a combination of training-aware neural architecture search and scaling, to jointly optimize training speed and parameter efficiency. The models were searched from the search space enriched with new ops such as Fused-MBConv. Our experiments show that EfficientNetV2 models train much faster than state-of-the-art models while being up to 6.8x smaller.Our training can be further sped up by progressively increasing the image size during training, but it often causes a drop in accuracy. To compensate for this accuracy drop, we propose an improved method of progressive learning, which adaptively adjusts regularization (e.g. data augmentation) along with image size.With progressive learning, our EfficientNetV2 significantly outperforms previous models on Im-ageNet and CIFAR/Cars/Flowers datasets. By pretraining on the same ImageNet21k, our Effi-cientNetV2 achieves 87.3% top-1 accuracy on ImageNet ILSVRC2012, outperforming the recent ViT by 2.0% accuracy while training 5x-11x faster using the same computing resources. Code is available at https://github.com/google/ automl/tree/master/efficientnetv2.
translated by 谷歌翻译