yoohj0416
commented
8 months ago Tag classification with one image
We did tag classification with one image per video. We used validation images of BDD100K (they don't provide test labels) and compared with accuracy in the paper. DLA-34 represents results from the Berkeley Deep Drive team (not sure either they used validation set or test set).
Weather classification
Scene classification
- The accuracy of weather classification is overall lower than DLA-34.
- The accuracy of scene classification is higher or similar to DLA-34 for all categories except residential (13 / 1253).
- We need to analyze why classification of residential image is extremely low.
To-Do
This is a set of steps for tag classification only with image.
In the tag classification test with one image experiment, classification is performed using only one image per video. The image is selected by the Berkeley DeepDrive team and provided by BDD100K images dataset.
In the tag classification with multiple images experiment, multiple images are extracted from each video and their embedding vectors are calculated. The vectors are then combined (i.e. mean) to perform classification.
Updated To-Do