Japanese Image-Text Retrieval Embdding with CLIP

author: David Wang

Description

This operator extracts features for image or text with Japanese-CLIP developed by rinna Co., Ltd, which can generate embeddings for Japanese text and image by jointly training an image encoder and text encoder to maximize the cosine similarity.

Code Example

Load an image from path './teddy.jpg' to generate an image embedding.

Read the text 'スケートボードに乗っているテディベア。' to generate an text embedding.

Write the pipeline in simplified style:

import towhee

towhee.glob('./teddy.jpg') \
      .image_decode() \
      .image_text_embedding.japanese_clip(model_name='japanese-clip-vit-b-16', modality='image') \
      .show()

towhee.dc(["スケートボードに乗っているテディベア。"]) \
      .image_text_embedding.japanese_clip(model_name='japanese-clip-vit-b-16', modality='text') \
      .show()

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

import towhee

towhee.glob['path']('./teddy.jpg') \
      .image_decode['path', 'img']() \
      .image_text_embedding.japanese_clip['img', 'vec'](model_name='japanese-clip-vit-b-16', modality='image') \
      .select['img', 'vec']() \
      .show()

towhee.dc['text'](["スケートボードに乗っているテディベア。"]) \
      .image_text_embedding.japanese_clip['text','vec'](model_name='japanese-clip-vit-b-16', modality='text') \
      .select['text', 'vec']() \
      .show()

Factory Constructor

Create the operator via the following factory method

japanese_clip(model_name, modality)

Parameters:

​ model_name: str

​ The model name of CLIP. Supported model names:

• japanese-clip-vit-b-16
• japanese-cloob-vit-b-16

​ modality: str

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

Interface

An image-text embedding operator takes a towhee image or string as input and generate an embedding in ndarray.

Parameters:

​ data: towhee.types.Image (a sub-class of numpy.ndarray) or str

​ The data (image or text based on specified modality) to generate embedding.

Returns: numpy.ndarray

​ The data embedding extracted by model.