[SSD] Small object detection

Hi all,

I have a question regarding the configuration of SSD.
An interesting task for me is to fine-tuning the SSD_mobilenet_v1_coco_2017_11_17 with Bosch small traffic light dataset.

However, the default setting is to resize the image into 300 x 300 (image_resizer).
Here is the total loss during training.
It loss maintains around 6. (Please ignore the overlapping at 5000 steps, due to some re-launch trainign process.)

I think the trend of the total loss is okay.
However, when I stop around 12k and feed with the test dataset (around 90 images for a short try). There is nothing detected.

Personally, I have some doubts about this issue:

1. Maybe the small traffic lights are too small for SSD?
2. However, why the total loss curve displayed a correct "learning" process?

Can I simply change the config of image size into 512 x 512 or even larger value (1000 x 1000)?
Will this work correctly as well?

Regards,
Hao

Start with the SSD_mobilenet_v1_coco_2017_11_17 configuration, especially the image_resizer setting mentioned in the issue, and review how the Bosch small traffic light dataset is prepared for training and evaluation. Compare the 300 x 300 setup with the proposed larger sizes and verify whether detections appear on the test images; done means explaining or reproducing the missing detections and identifying a supported configuration.