FeyNoBg
Automatic background removal model and training library
Details
- External ID
- 49072462
- Source
- HN
- Company
- —
- Product
- FeyNoBg
- Website domain
- usefeyn.com
- Launched
- July 27, 2026
- Cohort
- —
- Upvotes
- 121
- Upvotes percentile
- 0.9396654719235364
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hey HN, I’m Shreyash from Feyn. We help companies build custom models from their data.Today, we’re releasing FeyNoBg, an automatic background removal model. Alongside it, we're open-sourcing NoBg, the Python library we built to train and run it.Try the model here: https://huggingface.co/spaces/feyninc/feynobg. Check out the library here: https://github.com/feyninc/nobgSome sample outputs:(1) Soccer Freekick: https://drive.google.com/file/d/1MZkAGLwbhNVOZ0Oi7XvpCfSEu9Q...(2) Hair in wind: https://drive.google.com/file/d/1Odc2m0XMVH9uZtvI_KjaRbXzhLL...(3) Bicycle with visible spokes: https://drive.google.com/file/d/1h99ahjfrtS1MFQJJgiKE2fuM3HZ...(4) Live Demo video: https://youtu.be/b1heHPvY8BMBackground removal separates an image's subject from its surrounding. We've all tried it at some point. Often it is to reuse the subject in a different artifact. Nowadays, it is common to make chat stickers out of it. It is one of the most common but under-appreciated uses of AI. It is also surprisingly complex. Models can be easily confused by camouflage, motion blur, or fine structures like hair.The task requires two skills. First, a model has to identify the foreground. Second, it has to trace the foreground’s boundary and estimate an opacity value for each pixel. Generally, these skills are taught with different datasets. That creates a failure point. A poor training mix can improve one skill at the expense of the other. We saw this in our controlled evaluation. A training run with just the MaskFactory dataset improved on the CAMO benchmark but regressed on DIS5K.For FeyNoBg, we took an interpretability-first approach to training. We first studied how BiRefNet’s stages contribute to finding the foreground and reconstructing its boundary. We discovered that the third stage of it's feature extractor holds a lot of information. Both localization and boundary reconstruction depend heavily on the feature map produced here.This led us to expand this stage from 18 to 24 blocks while preserving the pre-trained weights. We then trained FeyNoBg on 26.1K diverse examples assembled from 10 datasets. The goal was to improve foreground identification and boundary precision without sacrificing either one.Across eight benchmarks, FeyNoBg achieves the best published score on four and comes within 2% of the leader on the rest.Building FeyNoBg also exposed a tooling problem. Image matting models are usually released as isolated repositories with incompatible preprocessing, training, and evaluation code. We built NoBg to solve this. NoBg puts these workflows behind one Python interface. It supports BiRefNet today, with more architectures coming. We hope you build something exciting with it!Happy to answer any questions!
Enrichment
- Theme
- image editing and background removal tools
- Vertical
- Horizontal
- Function
- Dev tools
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- background removal model and training library
- Manually corrected
- False
Could you build this?
No Training a state-of-the-art vision model for high-resolution matting and background removal requires extensive deep learning expertise, curated proprietary datasets, and high-performance GPU clusters.
What it would actually take: Building this requires designing deep segmentation/matting architectures (e.g., modified ViT/U-Net backbones with alpha matte refinement), collecting and labeling hundreds of thousands of diverse images with pixel-perfect masks, and managing distributed multi-GPU training runs (PyTorch/CUDA). You also need low-level ONNX/TensorRT optimization to run inference fast in production. This requires dedicated computer vision research scientists and significant compute budget.
Discussion
20 comments analyzed.
Competitors mentioned: photoroom.com, Adobe background removal (Select Subject/Select Person), remove.bg, u2-net, s3od, PDFNet, BEN (permissively licensed alternatives)
Concerns raised: CC-BY-NC license creates legal questions for professional users, Automatic model can't distinguish what user wants as subject (person vs person+couch), Attribution shrinking despite actual usage increasing, Re-licensing MIT licensed derivative work (BiRefNet) raises IP concerns
Feature requests: Convert from automatic to prompt model for selecting specific foreground elements, Support for user-specified subject selection (like Adobe's Select Person)
Competitors
Other products that read as similar to this one — 41 launches clear the similarity bar, closest 8 shown.
Attention rank: #2 of 42 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 256 days after the earliest competitor.
- MultiMatte, a Promptable Image Background Removal Model · hn · 2026-09-10 · 57 upvotes · similarity 0.59
- RemoveBG Top 2.0 · ph · 2026-09-23 · 2 upvotes · similarity 0.45
- 73hi Background Remover · ph · 2026-09-22 · 1 upvotes · similarity 0.42
- WebRemoveBG · ph · 2026-09-25 · 1 upvotes · similarity 0.39
- nobg · ph · 2026-09-17 · 3 upvotes · similarity 0.39
- Purgebg · ph · 2026-09-11 · 2 upvotes · similarity 0.38
- Tensor Spy: inspect NumPy and PyTorch tensors in the browser, no upload · hn · 2026-03-02 · 22 upvotes · similarity 0.37
- LemonSlice · hn · 2026-01-27 · 133 upvotes · similarity 0.37
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a dev tools tool for Sales yet.