upload dataset file to repo

lisa_data/reasonseg.py

@@ -0,0 +1,147 @@
+import json
+import glob
+import random
+import cv2
+from tqdm import tqdm
+import numpy as np
+from pycocotools import mask as maskUtils
+
+def singleMask2rle(mask):
+    rle = maskUtils.encode(np.array(mask[:, :, None], order='F', dtype="uint8"))[0]
+    rle["counts"] = rle["counts"].decode("utf-8")
+    return rle
+
+
+def get_mask_from_json(inform, height, width):
+
+    ### sort polies by area
+    area_list = []
+    valid_poly_list = []
+    for i in inform:
+        label_id = i["label"]
+        points = i["points"]
+        if "flag" == label_id.lower():  ## meaningless deprecated annotations
+            continue
+
+        tmp_mask = np.zeros((height, width), dtype=np.uint8)
+        cv2.polylines(tmp_mask, np.array([points], dtype=np.int32), True, 1, 1)
+        cv2.fillPoly(tmp_mask, np.array([points], dtype=np.int32), 1)
+        tmp_area = tmp_mask.sum()
+
+        area_list.append(tmp_area)
+        valid_poly_list.append(i)
+
+    ### ground-truth mask
+    sort_index = np.argsort(area_list)[::-1].astype(np.int32)
+    sort_index = list(sort_index)
+    sort_inform = []
+    for s_idx in sort_index:
+        sort_inform.append(valid_poly_list[s_idx])
+
+    mask = np.zeros((height, width), dtype=np.uint8)
+    for i in sort_inform:
+        label_id = i["label"]
+        points = i["points"]
+
+        if "ignore" in label_id.lower():
+            label_value = 255  # ignored during evaluation
+        else:
+            label_value = 1  # target
+
+        cv2.polylines(mask, np.array([points], dtype=np.int32), True, label_value, 1)
+        cv2.fillPoly(mask, np.array([points], dtype=np.int32), label_value)
+
+    return mask
+
+SHORT_QUESTION_LIST = [
+    "Can you segment the {class_name} in this image?",
+    "Please segment the {class_name} in this image.",
+    "What is {class_name} in this image? Please respond with segmentation mask.",
+    "What is {class_name} in this image? Please output segmentation mask.",
+    "Could you identify and segment the {class_name} in this image?",
+    "Would you be able to segment the {class_name} in this image?",
+    "Can you provide a segmentation mask for the {class_name} in this image?",
+    "Please provide a segmentation mask for the {class_name} in this image.",
+    "Could you please segment the {class_name} in this image for me?",
+    "What {class_name} is present in this image? Kindly respond with a segmentation mask.",
+    "Which part of this image contains {class_name}? Please output with segmentation mask.",
+    "Is there a {class_name} in this image? If so, please provide the segmentation mask.",
+    "Can you segment out the {class_name} visible in this image?",
+    "Would you identify and provide a segmentation mask for the {class_name} in this image?",
+]
+
+LONG_QUESTION_LIST = [
+     "{sent} Please respond with segmentation mask.",
+     "{sent} Please output segmentation mask.",
+]
+
+EXPLANATORY_QUESTION_LIST = [
+    "Please output segmentation mask and explain why.",
+    "Please output segmentation mask and explain the reason.",
+    "Please output segmentation mask and give some explanation.",
+]
+
+ANSWER_LIST = [
+    "It is [SEG].",
+    "Sure, [SEG].",
+    "Sure, it is [SEG].",
+    "Sure, the segmentation result is [SEG].",
+    "[SEG].",
+]
+    
+
+json_data = glob.glob('/mnt/workspace/workgroup/yuanyq/code/LISA/dataset/reason_seg/ReasonSeg/train/*.json')
+final_data = []
+idx = 0
+for d in tqdm(json_data):
+    data = json.load(open(d))
+    texts = data['text']
+    img_path = d.replace('.json','.jpg')
+    image = cv2.imread(img_path)
+    h, w = image.shape[:2]
+    for text in texts:
+        dic = {}
+        dic['id'] = f'reasonseg_{idx}'
+        idx+=1
+        dic['image'] = img_path.replace('/mnt/workspace/workgroup/yuanyq/code/LISA/dataset/', '')
+        if data['is_sentence']:
+            question_template = random.choice(LONG_QUESTION_LIST)
+            question = question_template.format(sent=text)
+        else:
+            question_template = random.choice(SHORT_QUESTION_LIST)
+            question = question_template.format(class_name=text.lower())
+        
+        answer = random.choice(ANSWER_LIST)
+        dic['conversations'] = []
+        dic['conversations'].append({'from': 'human', 'value': '<image>\n' + question})
+        dic['conversations'].append({'from': 'gpt', 'value': answer})
+        dic['masks'] = []
+        msk = get_mask_from_json(data['shapes'], h, w)
+        rle = singleMask2rle(msk)
+        dic['masks'].append(rle)
+        # dic['height'] = h
+        # dic['width'] = w
+        final_data.append(dic)
+
+explanatory = json.load(open('/mnt/workspace/workgroup/yuanyq/code/LISA/dataset/reason_seg/ReasonSeg/explanatory/train.json'))   
+for data in tqdm(explanatory):
+    dic = {}
+    dic['id'] = f'reasonseg_{idx}'
+    idx+=1
+    dic['image'] = 'reason_seg/ReasonSeg/train/'+data['image']
+    dic['conversations'] = []
+    dic['conversations'].append({'from': 'human', 'value': '<image>\n' + data['query']})
+    dic['conversations'].append({'from': 'gpt', 'value': random.choice(ANSWER_LIST) + ' ' + data['outputs']})
+    json_data = json.load(open(f'/mnt/workspace/workgroup/yuanyq/code/LISA/dataset/reason_seg/ReasonSeg/train/{data["json"]}'))
+    dic['masks'] = []
+    msk = get_mask_from_json(json_data['shapes'], h, w)
+    rle = singleMask2rle(msk)
+    dic['masks'].append(rle)
+    image = cv2.imread('/mnt/workspace/workgroup/yuanyq/code/LISA/dataset/reason_seg/ReasonSeg/train/'+data['image'])
+    h, w = image.shape[:2]
+    dic['height'] = h
+    dic['width'] = w
+    final_data.append(dic)
+print(len(final_data)) 
+with open('reason_seg.json', 'w') as f:
+    f.write(json.dumps(final_data, indent=4))