RockYang/geekai

<div align="center"> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/images/logo.png" width = "200" /> </div> [](https://gitee.com/mymagicpower/AIAS/stargazers) [](https://gitee.com/mymagicpower/AIAS/members) <h4 align="center"> <p> <b>中文</b> | <a href="https://github.com/mymagicpower/AIAS/blob/main/README_EN.md">English</a> <p> </h4> <h4 align="center"> <p> <b>官网</b> | <a href="https://www.aias.top/">https://www.aias.top/</a> <p> </h4> </div> <br> <hr> <br> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/images/AIAS.png" width = "600" /> </div> <div align="center"> <table> <tr> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/images/aias_training1.png" width = "250" /> </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/images/aias_search.png" width = "250" /> </div> </td> </tr> <tr> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/image_search/images/face_search.png" width = "250" /> </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/train_platform/images/training.png" width = "250" /> </div> </td> </tr> </table> </div> #### 项目清单: - 1. 1_image_sdks - [图像识别 SDK] ```text 1). 工具箱系列：图像处理工具箱（静态图像） 2). 目标检测：目标检测、目标跟踪、人脸检测&识别 3). 图像分割：图像分割、医疗影像 4). 行为分析：行为识别、姿态估计 5). GAN： 超分辨率、动作驱动、风格迁移、图像生成 6). 其它类别：OCR、深度估计、视频理解、图像检索 ... ``` <div align="center"> <table> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 1：OCR方向检测与旋转 - ocr_sdks/ocr_direction_det_sdk</p> OCR图像预处理。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/RotationExample.jpeg" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 2：OCR文字识别 - ocr_sdks/ocr_v3_sdk</p> 1. V3 文本检测: <br> - 中文文本检测<br> - 英文文本检测<br> - 多语言文本检测<br> 2. V3 文本识别:<br> - 中文简体<br> - 中文繁体<br> - 英文<br> - 韩语<br> - 日语<br> - 阿拉伯<br> - 梵文<br> - 泰米尔语<br> - 泰卢固语<br> - 卡纳达文<br> - 斯拉夫<br> </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/OcrV3RecExample2.jpeg" width = "400px"/> </div> </td> </tr> <tr> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 3：OCR文字识别 - ocr_sdks/ocr_v4_sdk</p> 原生支持旋转倾斜文本文字识别。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/OcrV3RecExample2.jpeg" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 4：版面分析 - ocr_sdks/ocr_layout_sdk</p> 可以用于配合文字识别，表格识别的流水线处理使用。 <br> 1. 中文版面分析<br> 2. 英文版面分析<br> 3. 中英文文档 - 表格区域检测<br> </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/LayoutCNDetExample.jpeg" width = "400px"/> </div> </td> </tr> <tr> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 5： 表格识别 - ocr_sdks/ocr_table_sdk</p> 中英文表格识别。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/TableENRecExample.jpeg" width = "400px"/> </div> </td> </tr> <tr> <tr> <td style="width:220px"> <div align="left"> <p>OCR工具箱 6： led文字识别 - ocr_sdks/ocr_led_sdk</p> led表盘文字识别。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/OCR/images/led_rec_result.png" width = "400px"/> </div> </td> </tr> <tr> <td> <div align="left"> <p>动物分类识别 - classification/animal_sdk</p> 动物识别sdk，支持7978种动物的分类识别。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/animal_sdk/tiger.jpeg" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>菜品分类识别 - classification/dish_sdk</p> 菜品识别sdk，支持8416种菜品的分类识别。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/dish_sdk/dish.jpeg" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>单目深度估计 - depth_estimation_sdk</p> MiDaS_Small实现单目深度估计，模型可通过输入图像估计其中的深度信息。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/depth_estimation_sdk/depth.png" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>动作驱动 - first_order_sdk</p> sdk以一段动作视频去驱动一张图片运动。可以驱动任意类型的运动。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/gan_sdks/result.gif" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>实例分割 - instance_segmentation_sdk</p> 支持对图片中80个分类的目标进行实例分割。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/Instance_segmentation_sdk/result.jpeg" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>人群密度检测 - crowd_sdk</p> 统计人数，计算密度图。 </div> </td> <td> <div align="center"> <img src="https://aias-home.oss-cn-beijing.aliyuncs.com/AIAS/sec_sdks/images/density.png" width = "400px"/> </div> </td> </tr> <tr> <td style="width:220px"> <div align="left"> <p>烟�