[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![ai_for_success Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1618975370488999936.png) AshutoshShrivastava [@ai_for_success](/creator/twitter/ai_for_success) on x 63.9K followers
Created: 2025-07-22 00:32:00 UTC

Google introduces conversational image segmentation with Gemini XXX

> Gemini can segment objects in images using natural language no predefined labels needed
> Understands complex phrases like "the car that is farthest away"
> Supports multi-lingual queries and in-image text (OCR)
> Handles abstract concepts like “damage” or “mess”
> Processes logic-based queries like “people who are not sitting”

Best Practices:

> Use `gemini-2.5-flash` model
> Set `thinkingBudget=0`
> Use clear prompts and request output in JSON
> Each mask should return: `box_2d`, `mask`, `label`

![](https://pbs.twimg.com/ext_tw_video_thumb/1947454532772106244/pu/img/GeWuJkzTOQliQaJY.jpg)

XXXXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1947454544557875343/c:line.svg)

**Related Topics**
[abstract](/topic/abstract)
[$googl](/topic/$googl)
[stocks communication services](/topic/stocks-communication-services)

[Post Link](https://x.com/ai_for_success/status/1947454544557875343)