[英]Can ssd mobilenet v1 in object detection tensorflow api be tried with different resize shapes than the default ones?
In ssd_mobilenet_v1_coco.config the image_resizer default size is 300x300, or 512x512. 在ssd_mobilenet_v1_coco.config中,image_resizer的默认大小为300x300或512x512。 State of art results are available for the options only. 现状结果仅适用于选项。
But resizing to smaller sizes leads to information loss, can ssd mobilenet be tried with say size 720x720? 但是调整到较小的尺寸会导致信息丢失,ssd mobilenet可以尝试使用720x720的尺寸吗?
Config file: https://github.com/tensorflow/models/blob/master/research/object_detection/samples/configs/ssd_mobilenet_v1_coco.config 配置文件: https : //github.com/tensorflow/models/blob/master/research/object_detection/samples/configs/ssd_mobilenet_v1_coco.config
It depends whether you're asking about training or inference. 这取决于您是否询问培训或推理。
If your goal is to detect objects using a pre-trained model, then it is not recommended to change the resizing parameters, as the model is tuned to work best of these. 如果您的目标是使用预先训练的模型检测对象,则不建议更改调整大小参数,因为模型已调整为最佳。
However, if you wish to train the model, then yes, you can modify them. 但是,如果您希望训练模型,那么是的,您可以修改它们。 However, be aware that changing these values non-marginally would also require you to change the architecture and/or anchor configuration a bit, depending on the objects' sizes you wish to detect. 但是,请注意,非边缘地更改这些值还需要您稍微更改架构和/或锚点配置,具体取决于您要检测的对象大小。 For example, if you're using larger input resolution, than I would recommend adding SSD layers ( this is the original configuration, with 6 feature maps with stride of 8, 16, 32, 64, 128 and 256) and changing anchor scales ( this is the original, with 6 layers and linear scales in the range of 0.2-0.95 of the image input size). 例如,如果你使用更大的输入分辨率,那么我建议添加SSD图层( 这是原始配置,有6个特征图,步幅为8,16,32,64,128和256)并且更改锚点尺度( 这是原始的,具有6层和线性刻度,在图像输入尺寸的0.2-0.95范围内)。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.