Update README.md

README.md

@@ -67,29 +67,29 @@ Large-Scale 1080P Multi-View Human Motion Data for AI Research & Commercial Appl
       📂 Dataset Structure
     </h2>
     <p style="font-size:18px; color:#555; line-height:1.8; margin-bottom:15px; margin-top:0;">
-      数据集采用标准化的文件夹结构，便于快速加载和处理。每个动作序列包含多视角视频、同步时间戳及元数据。
+      We provide modularly selectable annotation data covering human body parameters ranging from video-level to frame-level.
     </p>
     <pre style="background:#f8f9fa; padding:18px; border-radius:10px; font-size:16px; color:#333; overflow-x:auto; border-left:4px solid #D72222; font-family: 'Courier New', monospace; line-height:1.6; margin:0 0 15px 0;">
 ├── Dataset_Root/
-│ ├── Videos/       # 原始视频数据
-│ │ ├── Duration/     # 时长
-│ │ ├── Scene_01/     # 一个完整采集单元
+│ ├── Videos/       # raw video data
+│ │ ├── Duration/     # duration
+│ │ ├── Scene_01/     # a complete acquisition unit
 │ │ │ ├── Cam_01.mp4
 │ │ │ └── ... (6 views)
-│ ├── Content/       # 场景内容描述
-│ │ ├── Environment/   # 环境
-│ │ ├── Scene_Type/    # 场景
-│ │ ├── People_Count/   # 人数
-│ │ ├── Action_Category/ # 动作
-│ ├── VGGT/         # 3D点云图
-│ ├── Annotations/     # SMPL 参数
+│ ├── Content/       # scene content description
+│ │ ├── Environment/   # environment
+│ │ ├── Scene_Type/    # scene
+│ │ ├── People_Count/   # number of people
+│ │ ├── Action_Category/ # action
+│ ├── VGGT/         # 3D Point Cloud
+│ ├── Annotations/     # SMPL Parameters
 │ │ ├── Scene_01.json
 │ │ │ ├── "frames": [
 │ │ │ │ ├──{
 │ │ │ │ │ ├── "frame_id": 0,
 │ │ │ │ │ ├── "smpl_params": {
 │ │ │ │ │ │ ├── "global_orient": [rx, ry, rz],
-│ │ │ │ │ │ ├── "body_pose": [[j1_rx, j1_ry, j1_rz],...], # 共23个关节
+│ │ │ │ │ │ ├── "body_pose": [[j1_rx, j1_ry, j1_rz],...], # 23 joints in total
 │ │ │ │ │ │ ├── "transl": [tx, ty, tz],
 │ │ │ │ │ ├── }
 │ │ │ │ ├──}
@@ -97,8 +97,8 @@ Large-Scale 1080P Multi-View Human Motion Data for AI Research & Commercial Appl
     </pre>
     <div style="background:#e3f2fd; padding:15px; border-radius:10px; margin-top:0; border-left:4px solid #2196F3;">
       <p style="margin:0; color:#1976D2; font-size:16px;">
-        <strong>💡 提示：</strong> 
-        这种结构遵循了计算机视觉领域的通用规范，可直接配合 PyTorch 或 TensorFlow 的 Dataset 类进行加载。
+        <strong>💡 Tip：</strong> 
+        Required fields can be freely chosen based on actual needs: multi-camera raw video data, environment, scene type, number of people, action category, duration, point cloud data (.vgt), frame-level human body parameters (SMPL).
       </p>
     </div>
   </div>
@@ -109,14 +109,14 @@ Large-Scale 1080P Multi-View Human Motion Data for AI Research & Commercial Appl
       📊 Dataset Overview
     </h2>
     <ul style="font-size:18px; color:#555; line-height:1.8; list-style-type:none; padding-left:0; margin:0 0 18px 0;">
-      <li style="margin-bottom:6px;">✅ <strong>多人动作：</strong> 800+ 小时</li>
-      <li style="margin-bottom:6px;">✅ <strong>双人动作：</strong> 2,200+ 小时</li>
-      <li style="margin-bottom:6px;">✅ <strong>单人动作：</strong> 7,000+ 小时</li>
-      <li style="margin-bottom:6px;">✅ <strong>真实场景数量：</strong> 2,000+</li>
-      <li style="margin-bottom:6px;">✅ <strong>参与拍摄人次：</strong> 6,000+</li>
-      <li style="margin-bottom:6px;">✅ <strong>实拍总时长：</strong> 10,000+ 小时</li>
-      <li style="margin-bottom:6px;">✅ <strong>视频质量：</strong> 全量 1080P 高清视频，单条最长视频超过 60 分钟</li>
-      <li style="margin-bottom:6px;">✅ <strong>采集方式：</strong> 多视角同步采集，时序稳定对齐</li>
+      <li style="margin-bottom:6px;">✅ <strong>multi-person actions：</strong> 800+ 小时</li>
+      <li style="margin-bottom:6px;">✅ <strong>two-person actions：</strong> 2,200+ 小时</li>
+      <li style="margin-bottom:6px;">✅ <strong>single-person actions：</strong> 7,000+ 小时</li>
+      <li style="margin-bottom:6px;">✅ <strong>number of real-world scenes：</strong> 2,000+</li>
+      <li style="margin-bottom:6px;">✅ <strong>Shooting participant person-times：</strong> 6,000+</li>
+      <li style="margin-bottom:6px;">✅ <strong>total actual shooting duration：</strong> 10,000+ 小时</li>
+      <li style="margin-bottom:6px;">✅ <strong>video quality：</strong> Full set of 1080P HD videos, with the longest single video exceeding 60 minutes.</li>
+      <li style="margin-bottom:6px;">✅ <strong>acquisition method：</strong> Multi-view synchronous acquisition, with stable time-series alignment.</li>
     </ul>
     <hr style="border:none; border-top:1px solid #eee; margin:0 0 12px 0;">
     <p style="font-size:18px; color:#555; margin:0; text-align:center;">