AD Research Hub
DashboardPeoplePapersWorkshopsDatasetsResearch Map
Login
⌘K
AD Research Hub — Anomaly Detection in Computer Vision
← Back to People

Jiajun Wu

Stanford University

Top CV ResearchersFrontier Research MapScore: 9h-index: 366 citations
HomepageHomepageTalks section on homepageTalk videoSemantic Scholar
Top CV Researcher — Rank #15 (top 10_20)

Assistant Professor

Contributions

3D perception, physical reasoning, generative modeling, vision-language reasoning

Why Selected

A leading younger vision researcher in 3D understanding, physical reasoning, and vision-language reasoning.

Score Breakdown

2

historical impact

3

recent visibility

2

current influence

2

asset availability

9

total

Frontier Research Map

Featured Work

Building Physically Plausible World Models

official workshop invited talk — 2025-07-18

Why Now

He is one of the clearest younger voices connecting computer vision, cognition, world models, and physical reasoning into one research program.

Key Ideas

  • -Visual intelligence needs structured representations of objects, relations, dynamics, and affordances rather than only dense embeddings.
  • -World models become more useful when they support reasoning, intervention, and physical plausibility instead of pure video realism.
  • -Neuro-symbolic and physics-aware ideas are returning because end-to-end scaling alone does not solve causal scene understanding.

Open Questions

  • ?What abstractions should a vision system expose if it must reason about the physical world instead of only describe it?
  • ?Can structured world models scale to open-ended internet data without losing interpretability?
  • ?Where is the right interface between symbolic structure and large generative models?
Younger Agenda-Setters and Adjacent ML Thinkershigh confidence
Cross-References

Themes

physical scene understandingworld modelsreasoningvisual reasoning

Sign in to access this content

Sign in