NSS Background Remover Platform vs Robust Video Matting

NSS Background Remover Platform is a free browser suite for removing backgrounds, editing images and video, and upscaling images. Image removal accepts PNG, JPG, WebP, AVIF, and HEIC files, handles batches of up to 100 images, and exports straight-alpha transparency. Video removal can isolate people or process other subjects frame by frame, producing transparent WebM or composited MP4 while keeping audio. Image tools include erase and restore brushes, a magic wand, edge refinement, background replacement, layers, and 3D mode. Video tools include scenes, object removal, grading, text overlays, layers, trimming, and export. The platform lists 40 free tools, 32 cinematic presets, four alpha-capable formats, and 66 help articles. Processing occurs inside the browser; NSS says files stay on the device and are not sent to its servers. No account, quota, watermark, or installation is required. Supported browsers include current listed versions of Chrome, Edge, Opera, Firefox, Safari, and iOS Safari. Limits include a 4096-pixel working side, 100 MB images, 500 MB videos, and duration caps.

Robust Video Matting is a free project for separating people from video backgrounds. It processes frames with a recurrent neural network that carries temporal memory, and the project says it can perform matting in real time without additional inputs. A browser webcam demo runs the model live and can visualize its recurrent states. The conversion API accepts a video or image sequence and can produce a composited video or PNG sequence, optionally including alpha and foreground outputs.

Models are available for PyTorch, TorchHub, TorchScript, ONNX, TensorFlow, TensorFlow.js, and CoreML. MobileNetv3 is recommended for most uses; ResNet50 is larger and described as offering small performance improvements. The README reports GPU throughput figures, including 4K at 76 FPS and HD at 104 FPS on a GTX 1080 Ti, and 4K at 154 FPS and HD at 172 FPS on an RTX 3090 with FP16 under specified settings. The project cautions that file conversion runs much slower because its script does not use hardware video encoding or decoding or parallel tensor transfers. CoreML models require iOS 13 or later and do not support dynamic resolution. The code is GPL-3.0 licensed.