SegmentLens is an independent AI image segmentation platform powered by the SAM 3 model and Google Gemini APIs, enabling users to remove, replace, and generate image content via text or visual prompts with pixel-level precision.
What is SegmentLens?
SegmentLens is an image segmentation tool that takes an uploaded image and a text or visual prompt, then outputs a precise segmentation mask and optionally a regenerated image (background replacement, style transfer, etc.). It runs on the SAM 3 model for segmentation and Google Gemini for generation, and is developed by an independent team (not affiliated with Meta or Google).
Key Features
- SAM 3 segmentation — Pixel-level object detection from text prompts, visual prompts (clicks/boxes), and exemplar prompts, achieving 98%+ accuracy on portraits, products, backgrounds, and text regions.
- Multi-model generation — After segmentation, Gemini handles background replacement, style transfer, and creative generation; one upload yields multiple outputs.
- API integration — Standardized HTTP API requires three lines of code, returns segmentation masks in seconds.
- Video object tracking — SAM 3 tracks segmented objects across video frames, handling motion, occlusion, and lighting changes.
- Interactive refinement — Follow-up prompts (click, text, or box) correct mask errors without restarting.
- Free credits for new users — Claim free credits upon sign-in to start segmenting immediately.
Who is it for?
- E-commerce teams — Remove backgrounds or swap product scenes in seconds for catalog photos, no design skills needed.
- Educators and content creators — Isolate subjects for instructional images, generate custom visuals with text prompts.
- Developers — Integrate the HTTP API into websites or apps for automated segmentation and generation workflows.
What can you do with SegmentLens?
- Background removal and replacement — Upload a product photo, prompt with "white studio background", get a clean cutout on a new scene.
- Multi-object segmentation — Use a single text prompt like "dog, cat, person" to generate masks for all matching objects in one image.
- Text-driven image generation — After SAM 3 segments an object, Gemini generates a new background or style based on your description.
How does SegmentLens work?
- Upload an image (or drag and drop) and optionally add text or visual prompts.
- SAM 3 analyzes the image and produces segmentation masks for all requested objects (seconds).
- Optionally, Gemini generates a new background or styled version from the mask.
Pricing
Freemium model. Paid tiers (Basic, Pro, Max) offer monthly credits (300, 600, 1500 respectively) with faster cloud processing and priority support. New users receive free credits upon sign-in.
FAQ
Is SegmentLens free?
New users get free credits after logging in. Continued use requires a paid plan (Basic, Pro, or Max) or you can stay within the free tier's limits.
Is SegmentLens affiliated with Meta or Google?
No. SegmentLens is an independent platform that uses SAM 3 (Meta) and Gemini (Google) APIs. It is not sponsored or endorsed by either company.
How does SAM 3 differ from SAM 2?
SAM 3 adds text-prompt segmentation, exemplar prompts, and video tracking — SAM 2 only supports click/box prompts and lacks text-driven object detection.
Can I use SegmentLens on mobile devices?
Yes. A dedicated iOS/macOS app is available for download from the App Store.
