mobilenetv3 ssd检测实战

1、Mobilenetv3（Searching for MobileNetV3）

    论文发出于2019年5月，作者google。

    论文链接：https://arxiv.org/pdf/1905.02244.pdf

    代码：https://paperswithcode.com/paper/searching-for-mobilenetv3

    摘要：

    We present the next generation of MobileNets based on a combination of complementary search techniques as well as a novel architecture design. MobileNetV3 is tuned to mobile phone CPUs through a combination of hardwareaware network architecture search (NAS) complemented by the NetAdapt algorithm and then subsequently improved through novel architecture advances. This paper starts the exploration of how automated search algorithms and network design can work together to harness complementary approaches improving the overall state of the art. Through this process we create two new MobileNet models for release: MobileNetV3-Large and MobileNetV3-Small which are targeted for high and low resource use cases. These models are then adapted and applied to the tasks of object detection and semantic segmentation. For the task of semantic segmentation (or any dense pixel prediction), we propose a new efficient segmentation decoder Lite Reduced Atrous Spatial Pyramid Pooling (LR-ASPP). We achieve new state of the art results for mobile classification, detection and segmentation. MobileNetV3-Large is 3.2% more accurate on ImageNet classification while reducing latency by 20% compared to MobileNetV2. MobileNetV3-Small is 6.6% more accurate compared to a MobileNetV2 model with comparable latency. MobileNetV3-Large detection is over 25% faster at roughly the same accuracy as MobileNetV2 on COCO detection. MobileNetV3-Large LRASPP is 34% faster than MobileNetV2 R-ASPP at similar accuracy for Cityscapes segmentation.

      注：如果看到太长的英文，一遍读不懂，可以分析一下句子的组成结构，较快的摘取出要表达的意思。

      理解全文参考：https://cloud.tencent.com/developer/article/1467101

 

2、ssd（Single Shot MultiBox Detector）

    论文发出于2015年12月，作者

    论文链接：https://arxiv.org/pdf/1512.02325.pdf

    代码：https://github.com/weiliu89/caffe/tree/ssd

    摘要：

    We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple feature maps with different resolutions to naturally handle objects of various sizes. Our SSD model is simple relative to methods that require object proposals because it completely eliminates proposal generation and subsequent pixel or feature resampling stage and encapsulates all computation in a single network. This makes SSD easy to train and straightforward to integrate into systems that require a detection component. Experimental results on the PASCAL VOC, MS COCO, and ILSVRC datasets confirm that SSD has comparable accuracy to methods that utilize an additional object proposal step and is much faster, while providing a unified framework for both training and inference. Compared to other single stage methods, SSD has much better accuracy, even with a smaller input image size. For 300×300 input, SSD achieves 72.1% mAP on VOC2007 test at 58 FPS on a Nvidia Titan X and for 500×500 input, SSD achieves 75.1% mAP, outperforming a comparable state of the art Faster R-CNN model.

 

3、mobilenetv3-ssd代码分析及实验

    代码链接：https://github.com/shaoshengsong/MobileNetV3-SSD 和  https://github.com/tongyuhome/MobileNetV3-SSD

    两者结合的原理：将MobilenetV3抽出二层特征层+另外补充的4层卷积层输出的特征，进行分类和回归。

    实践需要注意的理解：代码使用 source_layer_indexes = [GraphPath(11, 'conv', -1),

                                                                                  # 11,

                                                                                   20] 是mobilenetv3-small中使用特征层的信息，

                       

                                              source_layer_indexes = [GraphPath(16, 'conv', -1),

                                                                                 # 11,

                                                                                 22] 是mobilenetv3-large中使用特征层的信息。

       实践改进：此时ssd特征层分支的第一层特征图size是19*19，如果要将ssd的特征层输出大小变为38*38，19*19的话，以mobilenetv3-large为例，设置为：

                      

                      source_layer_indexes = [GraphPath(10, 'conv', -1), GraphPath(16, 'conv', -1),

                                                         # 11,

                                                         22] ，然后需要修改classifficition和regression部分的通道数。