03 of 04

Machines that move

Video

When the model adds time, what starts to drift?

PoseNet can locate keypoints in one frame; CoTracker can follow selected points across frames. Generative video faces the harder inverse problem: creating each next frame while preserving coherence in identity, setting, and motion.

A.

What we make visible

01

Frame

A video begins as still images ordered in time.

02

Keypoints / features

PoseNet turns one frame into estimated body keypoints; other models represent edges, texture, objects, or regions.

03

Correspondence across frames

CoTracker follows selected points through a sequence, making “where did this feature go?” a concrete computational question.

04

Coherence or drift

A generator must create new frames while preserving identity, layout, motion, and physics; failures reveal where that relation breaks.

Teaching move · PoseNet → CoTracker: PoseNet can locate an elbow in one frame; CoTracker asks where that selected point went in later frames. Try that distinction in the Point Correspondence Lab, then name the harder generative problem: creating each new frame while keeping the elbow, the person, and the scene mutually coherent. Tracking analyzes correspondence—it is not itself video generation.

Fig. 03B — Drift stripsame subject, unstable time
B.

Choose a pathway

Use

Run a video test

Build the same five-frame motion idea twice in Coherence Animator, changing only which visual references remain available.

Observe / Critique

Label failures

Analyze pre-generated clips or still sequences for drift, physics breaks, and camera jumps.

Teach / Design

Adapt the A/B animation comparison

Use Coherence Animator, paper, whiteboards, or still frames to compare previous-only with anchor-plus-previous drawing.

Build / Code

Make drift inspectable

Prototype a frame viewer, overlay, or annotation tool for temporal failure modes.

Critical / No-AI

Ask what should count

Critique video as evidence, consent object, classroom artifact, or synthetic record.

C.

Tools for this session

Go deeper

explore on your own · studio / async
D.

Reusable 120-minute facilitation plan

Plan, not historical record. The July 25 live session followed a different path: assignment reviews, Session 2 feedback, the Coherence Animator and tool discussion, and Dr. Emily Thomforde’s talk. See the recap and recording chapters above for what happened live.

0-12

Welcome + two artifact shares

Re-enter through the mechanism map and three-line camp argument. Respond first through a private note or poll, then hear two pre-arranged shares—including one classroom-facing example: one artifact, one claim, one boundary each.

12-24

Active synthesis + point correspondence

Retrieve the Session 1 and 2 findings through private writing, a poll or precise chat prompt, and two voluntary responses. Then predict and reveal one clean track; distinguish tracking existing frames from generating new ones.

24-47

Coherence Animator · solo A/B comparison

Draw one shared opening and four new frames per run—nine drawings total. Run A shows only the previous frame; Run B shows the shared opening anchor plus the previous frame. Choose one feature, play A twice and B twice at the same speed, then name what changed and what the activity cannot prove about a real model.

47-57

Curated failure hunt

Track one feature at a time across two frozen examples. Ask four reusable questions: what changed, what supports it, what would verify the source, and what can we not conclude?

57-65

System map

Map model, interface, training data, workers, prompter, subject, editor, and platform. Keep consent and provenance separate, and turn the four questions into a reusable classroom protocol.

65-83

One-tool studio

Start from a completed worked example. The default frozen route ends with an observation, exact evidence, bounded claim, missing source evidence, and one “With learners, I would…” sentence.

83-90

Two claims + guest introduction

Hear two concise participant claims, then introduce Dr. Emily Thomforde and the listening question.

90-116

Guest talk + Q&A

Protect 18 minutes for the guest talk and eight minutes for participant questions.

116-120

Guest takeaway + close

Ask for one sentence to carry forward, return to fluent ≠ true / plausible ≠ neutral / smooth ≠ evidence, and invite the optional Studio.

E.

Discussion prompts

What exactly changed?
What evidence supports that observation?
What would verify the clip’s source?
What can we not conclude?
A tracker follows evidence in existing frames. What is different about generating new frames?
How could you use these four questions with learners?

Low-AI / No-AI pathway

Participants can complete the session with Coherence Animator, curated clips, or the same nine-drawing A/B sequence on paper. Temporal Telephone remains an optional facilitator-led group version. Direct video generation is optional. Open the full No-AI pathway.