---
title: "Reconstructed Slides: Why your agents need decision traces, not just documents — Zach Blumenfeld, Neo4j"
category: "slides"
video_id: "B9h9ovW5H9U"
sourceLabels: ["Cropped public YouTube video frames", "Local OpenCV slide-region detection", "Local RapidOCR"]
---

# Reconstructed Slides: Why your agents need decision traces, not just documents — Zach Blumenfeld, Neo4j

## Source Video
[Why your agents need decision traces, not just documents — Zach Blumenfeld, Neo4j](https://www.youtube.com/watch?v=B9h9ovW5H9U)

## Method
This deck is reconstructed from the existing video frame captures by detecting likely slide regions with OpenCV, cropping/upscaling those regions, deduplicating similar crops, and OCRing the cropped slide images locally. It is a cleaner companion to the full-stage frame deck.

## Reconstructed Slides
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-001.jpg]]

- Source frame: `slide-001.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `157.8`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-002.jpg]]

- Source frame: `slide-002.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `166.23`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-003.jpg]]

- Source frame: `slide-003.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `177.64`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-004.jpg]]

- Source frame: `slide-004.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `164.49`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-005.jpg]]

- Source frame: `slide-005.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `165.93`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-006.jpg]]

- Source frame: `slide-006.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `176.62`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-007.jpg]]

- Source frame: `slide-007.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `177.69`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-008.jpg]]

- Source frame: `slide-008.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `176.89`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-009.jpg]]

- Source frame: `slide-009.jpg`
- Crop: `contour` `[0, 0, 960, 540]` score `177.18`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-010.jpg]]

- Source frame: `slide-010.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.87`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-011.jpg]]

- Source frame: `slide-011.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `177.22`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-012.jpg]]

- Source frame: `slide-012.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `166.36`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-013.jpg]]

- Source frame: `slide-013.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `164.33`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-014.jpg]]

- Source frame: `slide-014.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.19`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-015.jpg]]

- Source frame: `slide-015.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `176.8`
![[assets/reconstructed-slides/B9h9ovW5H9U/slide-016.jpg]]

- Source frame: `slide-016.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `166.79`
## Dense Scene-Detected Slide Candidates
- [[youtube-B9h9ovW5H9U-dense-slides]]