HZAI-ZJNU/Mamba-YOLO

the official pytorch implementation of “Mamba-YOLO：SSMs-based for Object Detection”

What it solves

It provides a simple baseline for object detection that leverages State Space Models (SSMs) instead of traditional architectures, aiming to maintain efficiency while achieving competitive detection accuracy on datasets like MSCOCO2017.

How it works

Mamba YOLO integrates the selective-scan mechanism (sourced from VMamba) into a YOLO-style object detection framework. It is built upon the Ultralytics codebase and offers three model scales: Tiny (T), Base (B), and Large (L), which vary in parameter count and computational cost (FLOPs).

Who it’s for

Computer vision researchers and developers looking for efficient object detection models that utilize State Space Models rather than standard convolutional or transformer-based backbones.

Highlights

  • State Space Model Integration: Uses a selective-scan approach for feature extraction.
  • Multiple Model Scales: Offers pre-trained versions (T, B, and L) to balance performance and resource constraints.
  • ** uma-based Implementation**: Developed using PyTorch and compatible with the Ultralytics ecosystem.

Related

  • Project
    state-spaces/mamba

    Mamba is a linear-time sequence modeling architecture based on selective state space models (SSMs) that provides a high-performance, subquadratic alternative to Transformers.

  • Project
    yzfzzz/depth-detect

    A high-performance C++ and TensorRT framework that fuses YOLO object detection with monocular/binocular depth estimation to track objects and estimate their motion states.

  • Project
    ultralytics/ultralytics

    A high-performance computer vision framework implementing YOLO models for real-time object detection, segmentation, classification, and pose estimation.

  • Project
    Tencent/YOLO-Master

    A YOLO-style framework for real-time object detection that uses Mixture-of-Experts (MoE) to adaptively allocate computation based on scene complexity for higher precision and lower latency.

  • Project
    ultralytics/yolo-ios-app

    A Swift-based SDK and demo app that brings real-time YOLO object detection and vision tasks to iOS devices using Core ML.