AI Engineer for Image Recognition
Budget: $250 – $750 AUD
Job Description:
We are looking for an experienced AI/ML Engineer to develop a highly efficient and accurate Virtual Try-On (VTO) model for our fashion app. The model should allow users to upload or scan clothing items and visualize how they look on a human model or themselves in real-time.
Key Responsibilities:
✅ Develop an AI-powered virtual try-on model capable of overlaying clothing onto a person’s image realistically.
✅ Ensure high accuracy in garment fit, texture preservation, and body alignment for a natural appearance.
✅ Optimize the model for fast inference speed (should work efficiently on mobile & web platforms).
✅ Implement pose estimation, segmentation, and warping techniques to ensure garments adapt to body movements.
✅ Fine-tune and train on fashion datasets to improve clothing fit and style recommendations.
✅ Integrate the model with a React Native-based frontend (optional but preferred).
✅ Ensure the AI adheres to real-world physics (e.g., garment draping, fabric behavior).
Ideal Candidate:
? Strong expertise in Computer Vision, Deep Learning, and AI-based Image Processing.
? Experience with State-of-the-Art Virtual Try-On Models (e.g., VITON-HD, HR-VITON, TryOnDiffusion).
? Proficiency in PyTorch, TensorFlow, OpenCV, and GAN-based image synthesis.
? Experience with 3D garment fitting & segmentation (optional but a plus).
? Strong skills in model optimization for mobile & web deployment.
? Familiarity with AWS, Google Cloud, or on-device AI inference for efficient performance.
? Ability to deliver a working proof-of-concept (PoC) within a defined timeline.
⚙️ Tech Stack (Preferred but not mandatory):
? Frameworks: PyTorch / TensorFlow
? Computer Vision Libraries: OpenCV, Detectron2, MediaPipe
? Try-On Models (Preferred): VITON-HD, HR-VITON, ClothFlow, Latent Diffusion Models
? Deployment: ONNX, TensorFlow Lite (for mobile optimization)
? Deliverables:
✔️ A working AI model that can generate virtual try-on results in real-time.
✔️ Model documentation & integration guide.
✔️ Assistance with deploying the model for testing (preferably in an API format).
✔️ Benchmark reports comparing accuracy, speed, and efficiency.
⏳ Timeline & Budget:
⏳ Expected Duration: 4-6 weeks for a PoC; full model refinement as needed.
? Budget: Open to discussion based on expertise & past experience.
? How to Apply:
? Share your portfolio, GitHub, or past projects related to virtual try-on / AI in fashion.
? Provide details on how you would approach model selection, dataset curation, and optimization.
? Bonus: If you have pre-trained models or demos, share links or videos.
Looking forward to finding an AI expert who can help bring next-gen virtual try-on technology to life!
We are looking for an experienced AI/ML Engineer to develop a highly efficient and accurate Virtual Try-On (VTO) model for our fashion app. The model should allow users to upload or scan clothing items and visualize how they look on a human model or themselves in real-time.
Key Responsibilities:
✅ Develop an AI-powered virtual try-on model capable of overlaying clothing onto a person’s image realistically.
✅ Ensure high accuracy in garment fit, texture preservation, and body alignment for a natural appearance.
✅ Optimize the model for fast inference speed (should work efficiently on mobile & web platforms).
✅ Implement pose estimation, segmentation, and warping techniques to ensure garments adapt to body movements.
✅ Fine-tune and train on fashion datasets to improve clothing fit and style recommendations.
✅ Integrate the model with a React Native-based frontend (optional but preferred).
✅ Ensure the AI adheres to real-world physics (e.g., garment draping, fabric behavior).
Ideal Candidate:
? Strong expertise in Computer Vision, Deep Learning, and AI-based Image Processing.
? Experience with State-of-the-Art Virtual Try-On Models (e.g., VITON-HD, HR-VITON, TryOnDiffusion).
? Proficiency in PyTorch, TensorFlow, OpenCV, and GAN-based image synthesis.
? Experience with 3D garment fitting & segmentation (optional but a plus).
? Strong skills in model optimization for mobile & web deployment.
? Familiarity with AWS, Google Cloud, or on-device AI inference for efficient performance.
? Ability to deliver a working proof-of-concept (PoC) within a defined timeline.
⚙️ Tech Stack (Preferred but not mandatory):
? Frameworks: PyTorch / TensorFlow
? Computer Vision Libraries: OpenCV, Detectron2, MediaPipe
? Try-On Models (Preferred): VITON-HD, HR-VITON, ClothFlow, Latent Diffusion Models
? Deployment: ONNX, TensorFlow Lite (for mobile optimization)
? Deliverables:
✔️ A working AI model that can generate virtual try-on results in real-time.
✔️ Model documentation & integration guide.
✔️ Assistance with deploying the model for testing (preferably in an API format).
✔️ Benchmark reports comparing accuracy, speed, and efficiency.
⏳ Timeline & Budget:
⏳ Expected Duration: 4-6 weeks for a PoC; full model refinement as needed.
? Budget: Open to discussion based on expertise & past experience.
? How to Apply:
? Share your portfolio, GitHub, or past projects related to virtual try-on / AI in fashion.
? Provide details on how you would approach model selection, dataset curation, and optimization.
? Bonus: If you have pre-trained models or demos, share links or videos.
Looking forward to finding an AI expert who can help bring next-gen virtual try-on technology to life!