Chinese-CLIP-ViT-Large-Patch14-336px

Introduction

This is the large-version of the Chinese CLIP, with ViT-L/14@336px as the image encoder and RoBERTa-wwm-base as the text encoder. Chinese CLIP is a simple implementation of CLIP on a large-scale dataset of around 200 million Chinese image-text pairs. For more details, please refer to our technical report https://arxiv.org/abs/2211.01335 and our official github repo https://github.com/OFA-Sys/Chinese-CLIP (Welcome to star! 🔥🔥)

Use with the official API

We provide a simple code snippet to show how to use the API of Chinese-CLIP to compute the image & text embeddings and similarities.

from PIL import Image
import requests
from transformers import ChineseCLIPProcessor, ChineseCLIPModel

model = ChineseCLIPModel.from_pretrained("OFA-Sys/chinese-clip-vit-large-patch14-336px")
processor = ChineseCLIPProcessor.from_pretrained("OFA-Sys/chinese-clip-vit-large-patch14-336px")

url = "https://clip-cn-beijing.oss-cn-beijing.aliyuncs.com/pokemon.jpeg"
image = Image.open(requests.get(url, stream=True).raw)
# Squirtle, Bulbasaur, Charmander, Pikachu in English
texts = ["杰尼龟", "妙蛙种子", "小火龙", "皮卡丘"]

# compute image feature
inputs = processor(images=image, return_tensors="pt")
image_features = model.get_image_features(**inputs)
image_features = image_features / image_features.norm(p=2, dim=-1, keepdim=True)  # normalize

# compute text features
inputs = processor(text=texts, padding=True, return_tensors="pt")
text_features = model.get_text_features(**inputs)
text_features = text_features / text_features.norm(p=2, dim=-1, keepdim=True)  # normalize

# compute image-text similarity scores
inputs = processor(text=texts, images=image, return_tensors="pt", padding=True)
outputs = model(**inputs)
logits_per_image = outputs.logits_per_image  # this is the image-text similarity score
probs = logits_per_image.softmax(dim=1)  # probs: [[0.0219, 0.0316, 0.0043, 0.9423]]

However, if you are not satisfied with only using the API, feel free to check our github repo https://github.com/OFA-Sys/Chinese-CLIP for more details about training and inference.

Results

MUGE Text-to-Image Retrieval:

Setup	Zero-shot				Finetune
Metric	R@1	R@5	R@10	MR	R@1	R@5	R@10	MR
Wukong	42.7	69.0	78.0	63.2	52.7	77.9	85.6	72.1
R2D2	49.5	75.7	83.2	69.5	60.1	82.9	89.4	77.5
CN-CLIP	63.0	84.1	89.2	78.8	68.9	88.7	93.1	83.6

Flickr30K-CN Retrieval:

Task	Text-to-Image						Image-to-Text
Setup	Zero-shot			Finetune			Zero-shot			Finetune
Metric	R@1	R@5	R@10	R@1	R@5	R@10	R@1	R@5	R@10	R@1	R@5	R@10
Wukong	51.7	78.9	86.3	77.4	94.5	97.0	76.1	94.8	97.5	92.7	99.1	99.6
R2D2	60.9	86.8	92.7	84.4	96.7	98.4	77.6	96.7	98.9	95.6	99.8	100.0
CN-CLIP	71.2	91.4	95.5	83.8	96.9	98.6	81.6	97.5	98.8	95.3	99.7	100.0

COCO-CN Retrieval:

Task	Text-to-Image						Image-to-Text
Setup	Zero-shot			Finetune			Zero-shot			Finetune
Metric	R@1	R@5	R@10	R@1	R@5	R@10	R@1	R@5	R@10	R@1	R@5	R@10
Wukong	53.4	80.2	90.1	74.0	94.4	98.1	55.2	81.0	90.6	73.3	94.0	98.0
R2D2	56.4	85.0	93.1	79.1	96.5	98.9	63.3	89.3	95.7	79.3	97.1	98.7
CN-CLIP	69.2	89.9	96.1	81.5	96.9	99.1	63.0	86.6	92.9	83.5	97.3	99.2

Zero-shot Image Classification:

Task	CIFAR10	CIFAR100	DTD	EuroSAT	FER	FGVC	KITTI	MNIST	PC	VOC
GIT	88.5	61.1	42.9	43.4	41.4	6.7	22.1	68.9	50.0	80.2
ALIGN	94.9	76.8	66.1	52.1	50.8	25.0	41.2	74.0	55.2	83.0
CLIP	94.9	77.0	56.0	63.0	48.3	33.3	11.5	79.0	62.3	84.0
Wukong	95.4	77.1	40.9	50.3	-	-	-	-	-	-
CN-CLIP	96.0	79.7	51.2	52.0	55.1	26.2	49.9	79.4	63.5	84.9

Citation

If you find Chinese CLIP helpful, feel free to cite our paper. Thanks for your support!

@article{chinese-clip,
  title={Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese},
  author={Yang, An and Pan, Junshu and Lin, Junyang and Men, Rui and Zhang, Yichang and Zhou, Jingren and Zhou, Chang},
  journal={arXiv preprint arXiv:2211.01335},
  year={2022}
}