test_vid.mp4

[deepConnectionism]

Where is test_vid.mp4？

[tancik]

The detector takes a video as one of the arguments. test_vid.mp4 is an example filename for the video. It should be modified to the name of your captured video.

[Charmve]

I also had like to know where should test_vid.mp4 be put.

because I met some debug errors, as following.

$ python3 detector.py \
  --detector_model detector_models/stegastamp_detector \
  --decoder_model saved_models/stegastamp_pretrained \
  --video test_vid.mp4 \
  --save_video out/test_out

test_vid.mp4
Traceback (most recent call last):
  File "detector.py", line 196, in <module>
    main()
  File "detector.py", line 79, in main
    f_height, f_width = frame.shape[0:2]
AttributeError: 'NoneType' object has no attribute 'shape'

Thanks. @tancik

[HanGod1]

Hello, is your problem solved? Do you know what kind of file should I put in the save_video? Also what kind of video should I take? Could I use camera on the computer screen or printed posters?

[Charmve]

## 修改前
self.stn_params = Sequential([
            Input(shape=(8, 400, 400, 3)),
            Conv2D(32, (3, 3), strides=2, activation='relu', padding='same'),
            Conv2D(64, (3, 3), strides=2, activation='relu', padding='same'),
            Conv2D(128, (3, 3), strides=2, activation='relu', padding='same'),
            Flatten(),
            Dense(128, activation='relu')
        ])
        
## 修改后
self.stn_params1 = Conv2D(32, (3, 3), strides=2, activation='relu', padding='same')
self.stn_params2 = Conv2D(64, (3, 3), strides=2, activation='relu', padding='same')
self.stn_params3 = Conv2D(128, (3, 3), strides=2, activation='relu', padding='same')
self.stn_params4 = Flatten()
self.stn_params5 = Dense(128, activation='relu')