{"id":26015507,"url":"https://github.com/javierkaiser9/rgb-d_dual_input_machine_learning_model","last_synced_at":"2026-04-28T21:35:53.437Z","repository":{"id":280504922,"uuid":"942221546","full_name":"JavierKaiser9/RGB-D_Dual_Input_Machine_Learning_Model","owner":"JavierKaiser9","description":"This Demo presents a machine learning-based steering module for sidewalk navigation . Using a dual-input EfficientNetV2 model, it processes RGB-D data from an Intel RealSense D415 to classify sidewalk scenarios and generate real-time steering commands. Optimized with OpenVINO","archived":false,"fork":false,"pushed_at":"2025-03-04T09:00:41.000Z","size":23127,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-03-04T10:19:45.742Z","etag":null,"topics":["efficientnet","efficientnetv2","machine-learning","numpy","opencv","openvino","python3","realsense-camera","tensorflow"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/JavierKaiser9.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-03-03T19:10:58.000Z","updated_at":"2025-03-04T09:00:44.000Z","dependencies_parsed_at":"2025-03-04T10:29:49.379Z","dependency_job_id":null,"html_url":"https://github.com/JavierKaiser9/RGB-D_Dual_Input_Machine_Learning_Model","commit_stats":null,"previous_names":["javierkaiser9/rgb-d_autonomous_navigation_demo","javierkaiser9/rgbd-dual_input_machine_learning_model","javierkaiser9/rgb-d_dual_input_machine_learning_model"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JavierKaiser9%2FRGB-D_Dual_Input_Machine_Learning_Model","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JavierKaiser9%2FRGB-D_Dual_Input_Machine_Learning_Model/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JavierKaiser9%2FRGB-D_Dual_Input_Machine_Learning_Model/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JavierKaiser9%2FRGB-D_Dual_Input_Machine_Learning_Model/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/JavierKaiser9","download_url":"https://codeload.github.com/JavierKaiser9/RGB-D_Dual_Input_Machine_Learning_Model/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":242138826,"owners_count":20078007,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["efficientnet","efficientnetv2","machine-learning","numpy","opencv","openvino","python3","realsense-camera","tensorflow"],"created_at":"2025-03-06T03:18:34.847Z","updated_at":"2025-10-12T10:35:41.748Z","avatar_url":"https://github.com/JavierKaiser9.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🚀 Sidewalk Navigation Demo  \n**RGB-D Fusion with EfficientNetV2**  \n\n## 🌟 Overview  \n\nThis repository demonstrates the **architecture and implementation of a dual-input, single-output deep learning model** for real-time **sidewalk navigation**. A key focus is to:  \n💡 Showcase the working principles of a **2-input, 1-output EfficientNetV2 model** for RGB-D fusion.  \n💡 Present the **performance of OpenVINO-optimized models**.\n\n\u003cimg src=\"https://github.com/JavierKaiser9/RGB-D_Dual_Input_Machine_Learning_Model/blob/master/realsense_415D_camera.png\" width=\"500\" alt=\"Realsense 415-D\" title=\"RGB-D Realsense 415-D camera\" /\u003e\n\n---\n\n## 🛠️ Key Technologies  \n**Deep Learning Framework**: TensorFlow for model development and training  \n**Model Architecture**: **Dual-Input, Single-Output EfficientNetV2**  \n   - Input: **RGB + Depth** (from Intel RealSense D415)  \n   - Output: **Steering Command** (turn left, right, or go straight)\n\n\u003cimg src=\"https://github.com/JavierKaiser9/RGB-D_Dual_Input_Machine_Learning_Model/blob/master/Model_Architecture.png\" width=\"500\" alt=\"Architecture\" title=\"Architecture\" /\u003e\n\n✅ **Hardware Acceleration**: **Intel OpenVINO 2023.2** for real-time inference  \n✅ **Depth Sensing**: **Intel RealSense D415** for **RGB-D fusion**  \n✅ **Performance Optimization**: Model converted to **OpenVINO IR format** for embedded deployment  \n✅ **Real-Time Execution**: Achieves a mean of **50 FPS** on an embedded system without GPU  \n✅ **Development Environment**:  \n   - **TensorFlow**: 2.10  \n   - **OpenCV (CV2)**: 4.8.0  \n   - **Python**: 3.10  \n✅ **Training Hardware for Demo Model**: **NVIDIA GeForce RTX 3050**  \n\n---\n\n#### 💻 Requirements  \n- **Intel RealSense D415** camera connected to your computer  \n- Python environment with **OpenVINO** installed\n\n## 🚀 How to Use  \n\nThis repository provides two main functionalities:  \n1️⃣ **Directly use the pre-trained OpenVINO model** for real-time inference.  \n2️⃣ **Train your own model** using the provided architecture.  \n\n### 🔹 1. Running the Pre-Trained OpenVINO Model  \nIf you want to use the **pre-trained OpenVINO model**, clone the repository and run the test_openvino_models.py file.\n\n### 🔹 2. Train your own OpenVINO Model  \nIf you want to train your own model, change the paths in the train_two_input_one_output_model.py file to the locations where you want to store the training and test data. Then, you can transform the TensorFlow model into an OpenVINO model using the create_openvino_model.py file.\n\n## 🎯 Performance Highlights  \n🔥 **High accuracy** in sidewalk scenario classification  \n⚡ Optimized for **low-latency execution** on edge devices  \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjavierkaiser9%2Frgb-d_dual_input_machine_learning_model","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjavierkaiser9%2Frgb-d_dual_input_machine_learning_model","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjavierkaiser9%2Frgb-d_dual_input_machine_learning_model/lists"}