Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

MultiMatte, a Promptable Image Background Removal Model

Details

External ID
49645803
Source
HN
Company
—
Product
MultiMatte, a Promptable Image Background Removal Model
Website domain
usefeyn.com
Launched
Sept. 10, 2026
Cohort
—
Upvotes
57
Upvotes percentile
0.8692185007974481
Tags
—
Fetched at
Sept. 14, 2026, 5:28 p.m.
Updated at
Sept. 14, 2026, 5:28 p.m.

Description

Hey HN, I'm Shreyash from Feyn. We help companies build custom models from their data.Today we're releasing MultiMatte, a background removal model you can aim with words. Name an object in your image. MultiMatte keeps just that thing, and removes everything else.Try it out on your images: https://usefeyn.com/multimatte/.Demo video: https://youtu.be/XZ5BJWAkOjsMultiMatte is open source. Build with it using our NoBg library https://github.com/feyninc/nobg. Model card here: https://hf.co/feyninc/multimatteMultiMatte is the second iteration of our background removal models. The first was FeyNoBg, which we released here https://news.ycombinator.com/item?id=49072462.The big upgrade is promptability. Most models keep all foreground elements when cutting the background. MultiMatte lets you prompt the exact objects you want to keep and remove everything else. For example, If you have an image with a dog and a bowl, you can ask MultiMatte to keep just the dog.MultiMatte is built on Meta's SAM 3, a concept-promptable detector that already understands phrases. Our biggest change was in masking. Instead of binary masks that classify each pixel as being inside or outside an object, MultiMatte uses alpha mattes that assign an opacity value to each pixel, with respect to an object. This allows us to better represent hair, fur, motion blur, and other kinds of fuzzy boundaries.Across all our measured benchmarks, MultiMatte shows a step improvement over SAM 3. On DIS5K, S-measure rises from 0.674 to 0.908 (a 34.6% relative gain), and on DUT-OMRON from 0.792 to 0.901 (13.7%).Our blog covers more of the training details and results https://usefeyn.com/blog/multimatte.Happy to answer any questions!

Enrichment

Theme
image editing and background removal tools
Vertical
Horizontal
Function
Content generation
Audience
Developer
AI stance
AI-native
Project type
Commercial product
Normalized one-liner
image background removal model
Manually corrected
False

Could you build this?

No This is a custom computer vision research project involving low-rank fine-tuning (LoRA) of SAM 3 (Segment Anything Model) for promptable alpha matting.

What it would actually take: The system requires a PyTorch-based training pipeline utilizing large-scale image matting and segmentation datasets with paired text annotations, running LoRA fine-tuning across high-end GPU clusters (A100/H100s). The hard part is adapting SAM's binary segmentation outputs into soft, continuous alpha trimaps/mattes without losing fine details like hair or translucency based on open-vocabulary text prompts. This requires specialized deep learning expertise in computer vision, matting loss functions, and model quantization/serving (e.g., TensorRT-LLM/vLLM for vision).

Discussion

8 comments analyzed.

Competitors mentioned: remove.bg (being shut down by Canva), Preview.app background removal tool

Concerns raised: Demo runs on AWS VM, not fully client-side

Feature requests: Video background removal capabilities, Optimized bounding box preprocessing, Outline tracing or contour hints as input, Support for markup image with visual hints as preprocessing

Competitors

Other products that read as similar to this one — 74 launches clear the similarity bar, closest 8 shown.

Attention rank: #5 of 75 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 301 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a content generation tool for Government yet.