Update README.md

README.md

@@ -2,16 +2,21 @@
 license: openrail
 ---
 
-# OVDEval：How to Evaluate the Generalization of Detection? A Benchmark for Comprehensive Open-Vocabulary Detection
+<h1 align="center"> OVDEval </h1>
+<h2 align="center"> A Comprehensive Evaluation Benchmark for Open-Vocabulary Detection</h2>
+<p align="center">
+ <a href="https://arxiv.org/abs/2308.13177"><strong> [Paper 📄] </strong></a>
+</p>
 
 ## Dataset Description
 
 OVDEval is a new benchmark for OVD model, which includes 9 sub-tasks and introduces evaluations on commonsense knowledge, attribute understanding, position understanding, object relation comprehension, and more. The dataset is meticulously created to provide hard negatives that challenge models' true understanding of visual and linguistic input. Additionally, we identify a problem with the popular Average Precision (AP) metric when benchmarking models on these fine-grained label datasets and propose a new metric called Non-Maximum Suppression Average Precision (NMS-AP) to address this issue.
 
 
-## Languages
+### Data Details
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/658a2e94991d8e7fb24f7688/ngOkek9wJdppyxPB0xZ8Q.png)
 
-The dataset contains questions in English and code solutions in Python.
 
 ## Dataset Structure
 
@@ -44,10 +49,10 @@ The dataset contains questions in English and code solutions in Python.
       "height": 254,
       "width": 340,
       "text": [
-        "computer without screen on"  # "text" represents the positive sample in this image.
+        "computer without screen on"  # "text" represents the annotated positive labels of this image.
       ],
       "neg_text": [
-        "computer with screen on" # "neg_text" represents the category for which this image does not belong.
+        "computer with screen on" # "neg_text" contains fine-grained hard negative labels which are generated according specific sub-tasks.
       ]
     }]
 }
@@ -58,11 +63,9 @@ The dataset contains questions in English and code solutions in Python.
 
 Reference https://github.com/om-ai-lab/OVDEval
 
+## Languages
 
-### Data Fields
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/658a2e94991d8e7fb24f7688/ngOkek9wJdppyxPB0xZ8Q.png)
-
+The dataset contains questions in English and code solutions in Python.
 
 ## Citation Information
 If you find our data, or code helpful, please cite the original paper: