aikenml/data_mining

1# Tutorial for WebUI 1.5 Version2## We have added two new features3- We have added text prompts to allow for interactive selection of objects that will be tracked in the video.4- We can now interactively add multiple objects for tracking in the video.5 6 7## Text-Prompts8### 1. Clone Grounding-DINO to `./src`9```10pip install -e git+https://github.com/IDEA-Research/GroundingDINO.git@main#egg=GroundingDINO11```12 13### 2. Switch to Text-Tab by clicking `Text` Tab14 15<p align="center">16<img src="./img/switch2textT.jpg" height="400">17</p>18 19### 3. Upload video or use example dicectly20 21### 4. Enter text to select the objects you are interested in22- The `.` is used to split text, just like in the original Grounding-Dino setting.23 24<p align="center">25<img src="./img/enter_text.jpg" height="400", width="400">26</p>27 28### 5. Get mask of selected object by clicking `Detect` button29- SAMTrack initialization may take some time.30 31<p align="center">32<img src="./img/detect_result.jpg" height="400", width="400">33</p>34 35### 6. Track in video36 37## Multi-Objects select38### 1. Once we interactively add an object mask, we can click the `Add new object button` to prepare to add a new object.39 40<p align="center">41<img src="./img/new_object.jpg" height="400", width="400">42</p>43 44### 2. Add a new object by clicking object45 46<p align="center">47<img src="./img/second_object.jpg" height="400", width="400">48</p>49 50### 3. You can add as many objects as you want by clicking `Add new object` button.