Inspiration
Understanding a dog’s emotions and health is often challenging because pets cannot communicate their feelings verbally. Many early signs of stress, anxiety, pain, or discomfort go unnoticed by owners, leading to delayed care and avoidable health issues. We were inspired to build PawSighnt to bridge this communication gap using AI, computer vision, and intelligent analysis, giving dogs a digital voice and helping owners better understand their pets’ emotional and physical well-being.
What it does
PawSighnt is an AI-powered veterinary emotion analyzer that interprets a dog’s emotions and behavior from video input. Users can upload a dog video and instantly view AI-generated insights including happiness level, stress indicators, activity level, and overall mental health estimation. The system also generates an AI vet summary in natural language and provides voice feedback using text-to-speech. In addition, an AI chatbot allows users to ask questions related to their dog’s behavior, emotions, or health and receive intelligent, vet-style responses.
How we built it
PawSighnt is built using a combination of modern AI technologies and a modular architecture. Flask is used to develop the web-based interface for video uploads and result visualization. YOLO is employed for pose estimation and keypoint detection, enabling the system to track tail, ear, head, and body posture. These keypoints are processed through custom behavior and emotion analysis modules to infer emotional states. DeepSeek and Novita are used for AI reasoning to generate human-readable veterinary insights, while ElevenLabs converts these insights into natural-sounding voice feedback. A hardware implementation is also developed using the RDK X5 kit with YOLO running on-device for real-time edge analysis.
Problem Statement & Motivation
Pet owners often misinterpret or overlook subtle behavioral changes in their dogs, which can be early signs of stress, discomfort, or health issues. Traditional monitoring methods rely heavily on human observation, which is subjective and inconsistent. There is a strong need for an intelligent, automated system that can continuously analyze dog behavior and emotions to provide meaningful insights. PawSighnt is motivated by the goal of improving animal welfare, enabling early intervention, and strengthening the bond between pets and their owners through technology.
Challenges we ran into
One of the major challenges was accurately mapping pose keypoints to meaningful emotional states. Dogs vary in posture, size, and movement patterns, which required careful logic design and calibration. Integrating multiple AI components such as computer vision, language models, and text-to-speech into a single smooth pipeline was also technically complex. Another challenge was ensuring real-time performance on hardware while maintaining accuracy. Designing a clean and intuitive UI that presents complex AI outputs in a simple way was also a significant challenge.
Accomplishments that we're proud of
We are proud of successfully building a complete end-to-end system that integrates computer vision, emotion analysis, AI reasoning, voice output, and both software and hardware implementations. The ability to upload a video, view emotion analysis, receive an AI vet summary, listen to voice feedback, and interact with a chatbot within a single interface is a major achievement.
What we learned
Through this project, we gained deep hands-on experience in integrating computer vision with AI reasoning and user interface design. We learned how to transform raw pose data into meaningful behavioral insights and how to design AI systems that are not only intelligent but also user-friendly. We also learned about the challenges of deploying AI on hardware and the importance of optimizing for performance and reliability. Most importantly, we learned how powerful AI can be when applied to real-world, emotionally meaningful problems.
What's next for PawSighnt
In the future, we plan to expand PawSighnt to support multiple animals such as cats and livestock, add mobile application support, and integrate wearable sensors for more accurate health monitoring.
Log in or sign up for Devpost to join the conversation.