package video_shot_sentences

Get desktop application:
View/edit binary Protocol Buffers messages

• optional int32 video_id = 1

  Video id required for paragraph models

• repeated int32 positive_shot_id = 2

  shot id of the positive word

• optional string video_name = 3

  Name of the video file

• repeated caffe.Datum positive_shot_words = 4

  The target shots which is to be predicted

• repeated caffe.Datum context_shot_words = 5

  The shot words which are in the context

• repeated caffe.Datum negative_shot_words = 6

  Negatives from same video

• repeated int32 negative_shot_id = 7

  shot id of the negative word

• optional int32 video_id = 1

  Video id required for paragraph models

• optional int32 shot_id = 2

  shot id of the word

• optional string video_name = 3

  Name of the video file

• optional caffe.Datum target_shot_word = 4

  The target shot which is to be predicted

• repeated caffe.Datum context_shot_words = 5

  The shot words which are in the context

• optional int32 video_id = 1

  Video id required for paragraph models

• repeated int32 shot_ids = 2

  shots ids for the shots

• repeated caffe.Datum shot_words = 3

  shots in the video

• optional string video_name = 4

  name of the video