yushihu commited on
Commit
de7d203
1 Parent(s): 4101b50

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -16,7 +16,8 @@ language:
16
  - en
17
 
18
  ---
19
- This is the repo for the paper [PromptCap: Prompt-Guided Task-Aware Image Captioning](https://arxiv.org/abs/2211.09699)
 
20
 
21
  We introduce PromptCap, a captioning model that can be controlled by natural language instruction. The instruction may contain a question that the user is interested in.
22
  For example, "what is the boy putting on?". PromptCap also supports generic caption, using the question "what does the image describe?"
@@ -43,7 +44,7 @@ Generate a prompt-guided caption by following:
43
  import torch
44
  from promptcap import PromptCap
45
 
46
- model = PromptCap("vqascore/promptcap-coco-vqa") # also support OFA checkpoints. e.g. "OFA-Sys/ofa-large"
47
 
48
  if torch.cuda.is_available():
49
  model.cuda()
@@ -87,7 +88,7 @@ import torch
87
  from promptcap import PromptCap_VQA
88
 
89
  # QA model support all UnifiedQA variants. e.g. "allenai/unifiedqa-v2-t5-large-1251000"
90
- vqa_model = PromptCap_VQA(promptcap_model="vqascore/promptcap-coco-vqa", qa_model="allenai/unifiedqa-t5-base")
91
 
92
  if torch.cuda.is_available():
93
  vqa_model.cuda()
 
16
  - en
17
 
18
  ---
19
+ This is the repo for the paper [PromptCap: Prompt-Guided Task-Aware Image Captioning](https://arxiv.org/abs/2211.09699). This paper is accepted to ICCV 2023 as [PromptCap: Prompt-Guided Image Captioning for VQA with GPT-3](https://openaccess.thecvf.com/content/ICCV2023/html/Hu_PromptCap_Prompt-Guided_Image_Captioning_for_VQA_with_GPT-3_ICCV_2023_paper.html).
20
+
21
 
22
  We introduce PromptCap, a captioning model that can be controlled by natural language instruction. The instruction may contain a question that the user is interested in.
23
  For example, "what is the boy putting on?". PromptCap also supports generic caption, using the question "what does the image describe?"
 
44
  import torch
45
  from promptcap import PromptCap
46
 
47
+ model = PromptCap("tifa-benchmark/promptcap-coco-vqa") # also support OFA checkpoints. e.g. "OFA-Sys/ofa-large"
48
 
49
  if torch.cuda.is_available():
50
  model.cuda()
 
88
  from promptcap import PromptCap_VQA
89
 
90
  # QA model support all UnifiedQA variants. e.g. "allenai/unifiedqa-v2-t5-large-1251000"
91
+ vqa_model = PromptCap_VQA(promptcap_model="tifa-benchmark/promptcap-coco-vqa", qa_model="allenai/unifiedqa-t5-base")
92
 
93
  if torch.cuda.is_available():
94
  vqa_model.cuda()