alimohammadiamirhossein/smite
Pytorch Implementation of "SMITE: Segment Me In TimE" (ICLR 2025)
SMITE helps you precisely identify and outline objects within video sequences, even if they're partially hidden or moving. You provide a video and a few example images with the objects already marked, and SMITE outputs the entire video with those objects segmented consistently across all frames. This is ideal for researchers or analysts working with video data who need to track specific items over time.
212 stars.
Use this if you need to consistently segment and track objects across many video frames with minimal manual effort, especially when dealing with occlusions or varying conditions.
Not ideal if your task involves static image segmentation only, or if you require real-time, ultra-low-latency video processing.
Stars
212
Forks
11
Language
Python
License
Apache-2.0
Category
Last pushed
Nov 12, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/alimohammadiamirhossein/smite"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hao-ai-lab/FastVideo
A unified inference and post-training framework for accelerated video generation.
ModelTC/LightX2V
Light Image Video Generation Inference Framework
thu-ml/TurboDiffusion
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
PKU-YuanGroup/Helios
Helios: Real Real-Time Long Video Generation Model
PKU-YuanGroup/MagicTime
[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators