Conversation
Co-authored-by: anxiangsir <31175974+anxiangsir@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
This PR adds visual documentation to the README by displaying three assets: a method overview diagram and two case demonstration GIFs.
- Added "Method Overview" section with architecture diagram showing the OneVision Encoder methodology
- Added "Case Demonstrations" section displaying two example cases (case4 and case6) in a side-by-side table layout
- Maintained consistency with existing image presentation patterns in the README
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| <img src="asset/case4.gif" alt="Case 4 Demonstration" width="400"><br> | ||
| <b>Case 4</b> | ||
| </td> | ||
| <td align="center"> | ||
| <img src="asset/case6.gif" alt="Case 6 Demonstration" width="400"><br> |
There was a problem hiding this comment.
The alt text for these case demonstrations is too generic and doesn't describe what the GIFs actually show. Good alt text should provide meaningful context about the content being displayed. Consider describing what each case demonstrates specifically (e.g., what task, behavior, or feature is being shown).
| <img src="asset/case4.gif" alt="Case 4 Demonstration" width="400"><br> | |
| <b>Case 4</b> | |
| </td> | |
| <td align="center"> | |
| <img src="asset/case6.gif" alt="Case 6 Demonstration" width="400"><br> | |
| <img src="asset/case4.gif" alt="Animated qualitative example for OneVision Encoder, Case 4 scenario" width="400"><br> | |
| <b>Case 4</b> | |
| </td> | |
| <td align="center"> | |
| <img src="asset/case6.gif" alt="Animated qualitative example for OneVision Encoder, Case 6 scenario" width="400"><br> |
| <tr> | ||
| <td align="center"> | ||
| <img src="asset/case4.gif" alt="Case 4 Demonstration" width="400"><br> | ||
| <b>Case 4</b> | ||
| </td> | ||
| <td align="center"> | ||
| <img src="asset/case6.gif" alt="Case 6 Demonstration" width="400"><br> | ||
| <b>Case 6</b> |
There was a problem hiding this comment.
The "Case Demonstrations" section lacks descriptive context. Unlike the "Input Method Comparison" section which includes a caption and explanatory text under each image, this section only shows "Case 4" and "Case 6" labels without explaining what these cases demonstrate or why they are important. Consider adding a table caption and brief descriptions to provide context for readers.
| <tr> | |
| <td align="center"> | |
| <img src="asset/case4.gif" alt="Case 4 Demonstration" width="400"><br> | |
| <b>Case 4</b> | |
| </td> | |
| <td align="center"> | |
| <img src="asset/case6.gif" alt="Case 6 Demonstration" width="400"><br> | |
| <b>Case 6</b> | |
| <caption style="caption-side: top; text-align: center; font-weight: bold; margin-bottom: 10px;"> | |
| Representative qualitative case studies showcasing OneVision Encoder’s behavior on diverse video understanding scenarios | |
| </caption> | |
| <tr> | |
| <td align="center"> | |
| <img src="asset/case4.gif" alt="Case 4 Demonstration: complex multi-event video understanding example" width="400"><br> | |
| <b>Case 4</b><br> | |
| Complex multi-event scene demonstrating temporal reasoning and motion-aware representation | |
| </td> | |
| <td align="center"> | |
| <img src="asset/case6.gif" alt="Case 6 Demonstration: long-duration activity and viewpoint change example" width="400"><br> | |
| <b>Case 6</b><br> | |
| Long-duration activity with viewpoint changes highlighting robust, sparse video encoding |
…thod-display Add method.jpg and case GIFs to README
Display three assets in README.md:
asset/method.jpg,asset/case4.gif, andasset/case6.gif.Changes
method.jpgarchitecture diagramcase4.gifandcase6.gifin side-by-side table layout (consistent with existing "Input Method Comparison" style)Original prompt
💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.