Describe your project

Project Description: Hashbied

• In-Scope:

  • Hashbied ensures the authenticity and integrity of video content by embedding cryptographic checksums and metadata into specific pixels.
  • It uses source identifiers, frame checksums, and validity bits to verify the integrity of each video frame during playback, detecting tampering or manipulation.
  • Supports high-resolution video frames (1920x1080p) and is designed for use in media production, content verification, and legal documentation workflows.
  • Encoding and decoding processes allow seamless integration into various video pipelines for secure and verifiable content distribution.

• Out of Scope:

  • Audio tampering detection and watermarking for intellectual property protection are not part of this solution.
  • Compression artifacts or post-production effects like color correction are not flagged as tampering, as they are part of legitimate editing.
  • Real-time tamper detection for live streaming is not supported at this stage; the focus is on pre-recorded video content.

• Future Opportunities:

  • Real-time tamper detection for live video streams (e.g., news broadcasts, sports events).
  • Audio verification integration, expanding Hashbied to an all-in-one multimedia verification tool.
  • Integration with cloud-based verification services and blockchain for decentralized content authentication.
  • Partnerships with content platforms to use Hashbied for authenticating user-generated content and combating misinformation.

Challenges we ran into

One of the major challenges we encountered was synchronizing the encoding and decoding processes for the video. The checksum calculation in our system excluded certain fixed "special pixels" used to store metadata, such as source identifiers and frame hashes. Initially, this led to discrepancies between the encoded and decoded data, resulting in numerous tampering flags for unaltered frames.

The challenge stemmed from ensuring that both the encoder and decoder consistently skipped these special pixels during checksum calculation. This desynchronization issue took considerable time to debug and fix, as we had to carefully align the logic in both processes to ensure the checksums matched, all while maintaining the integrity of the embedded metadata.

After thorough testing and refining, we finally synchronized the two processes, allowing for accurate tamper detection across the entire video.