RecallCam

Inspiration

I wanted to answer a simple question.What if a computer could remember what it saw?RecallCam was inspired by the everyday problem of forgetting where we left things.

What it does

RecallCam uses a webcam to detect, track, and remember objects. You can ask “Where is my phone?” or “What was I holding?” and it searches its visual memory.

How we built it

We used Python, YOLO, OpenCV, MediaPipe, SQLite, and voice interaction. Visual fingerprints help RecallCam recognize the same physical object after it moves.

Challenges we ran into

The hardest parts were reliable object re-identification, changing lighting, false detections, hand tracking, and maintaining real-time performance on limited hardware.

Accomplishments that we're proud of

We turned an ordinary webcam into a working prototype that doesn't just detect objects, it remembers them over time.

What we learned

We learned that detection is only the beginning. Building useful visual memory requires combining tracking, visual similarity, spatial information, and persistent storage.

What's next for RecallCam

We want to improve accuracy, long-term memory, natural-language queries, and eventually support multiple cameras.

Built With

• artificial
• assistant
• computer
• detection
• intelligence
• learning
• machine
• mediapipe
• memory
• natural-language-processing
• opencv
• python
• real-time
• recognition
• sqlite
• vision
• visual
• voice
• yolo