Video-Text Retrieval Embedding with Frozen In Time

author: Jinling Xu

Description

This operator extracts features for video or text with Frozen In Time which can generate embeddings for text and video by jointly training a video encoder and text encoder to maximize the cosine similarity.

Code Example

Load a video from path './demo_video.mp4' to generate a video embedding.

Read the text 'kids feeding and playing with the horse' to generate a text embedding.

Write the pipeline in simplified style:

• Encode video (default):

import towhee
towhee.dc(['./demo_video.mp4']) \
        .video_decode.ffmpeg(sample_type='uniform_temporal_subsample', args={'num_samples': 4}) \
        .runas_op(func=lambda x: [y for y in x]) \
        .video_text_embedding.frozen_in_time(model_name='frozen_in_time_base_16_244', modality='video', device='cpu') \
        .show()

• Encode text:

import towhee

towhee.dc(['kids feeding and playing with the horse']) \
      .video_text_embedding.frozen_in_time(model_name='frozen_in_time_base_16_244', modality='text', device='cpu') \
      .show()

Write a same pipeline with explicit inputs/outputs name specifications:

import towhee

towhee.dc['path'](['./demo_video.mp4']) \
        .video_decode.ffmpeg['path', 'frames'](sample_type='uniform_temporal_subsample', args={'num_samples': 4}) \
        .runas_op['frames', 'frames'](func=lambda x: [y for y in x]) \
        .video_text_embedding.frozen_in_time['frames', 'vec'](model_name='frozen_in_time_base_16_244', modality='video', device='cpu') \
        .show()

towhee.dc['text'](["kids feeding and playing with the horse"]) \
      .video_text_embedding.frozen_in_time['text','vec'](model_name='frozen_in_time_base_16_244', modality='text', device='cpu') \
      .select['text', 'vec']() \
      .show()

Factory Constructor

Create the operator via the following factory method

frozen_in_time(model_name, modality, weight_path)

Parameters:

​ model_name: str

​ The model name of frozen in time. Supported model names:

• frozen_in_time_base_16_244

​ modality: str

​ Which modality(video or text) is used to generate the embedding.

​ weight_path: str

​ pretrained model weights path.

Interface

An video-text embedding operator takes a list of Towhee VideoFrame or string as input and generate an embedding in ndarray.

Parameters:

​ data: List[towhee.types.Image] or str

​ The data (list of Towhee VideoFrame (which is uniform subsampled from a video) or text based on specified modality) to generate embedding.

Returns: numpy.ndarray

​ The data embedding extracted by model.