Join our in-person meetup on April 24th to hear talks from experts on cutting-edge topics across AI, ML, and computer vision. View more CV events here.

We present a novel large-scale dataset for defect detection in a logistics setting. Recent work on industrial anomaly detection has primarily focused on manufacturing scenarios with highly controlled poses and a limited number of object categories. Existing benchmarks like MVTec-AD (Bergmann et al., 2021) and VisA (Zou et al., 2022) have reached saturation, with state-of-the-art methods achieving up to 99.9% AUROC scores. In contrast to manufacturing, anomaly detection in retail logistics faces new challenges, particularly in the diversity and variability of object pose and appearance. Leading anomaly detection methods fall short when applied to this new setting. To bridge this gap, we introduce a new benchmark that overcomes the current limitations of existing datasets. With over 230,000 images (and more than 29,000 defective instances), it is 40 times larger than MVTec and contains more than 48,000 distinct objects. To validate the difficulty of the problem, we conduct an extensive evaluation of multiple state-of-the-art anomaly detection methods, demonstrating that they do not surpass 56.96% AUROC on our dataset. Further qualitative analysis confirms that existing methods struggle to leverage normal samples under heavy pose and appearance variation. With our large-scale dataset, we set a new benchmark and encourage future research towards solving this challenging problem in retail logistics anomaly detection. The dataset is available for download under https://www.kaputt-dataset.com.

At first glance, visual inspection of high-voltage power lines seems straightforward: collect imagery, run one or two AI models, and report the findings. In practice, moving beyond a proof of concept reveals a range of issues that can make or break a campaign. Common concerns include data quality and coverage, scarcity of the most relevant cases and abundance everywhere else, variations in pylon geometry and asset types across regions, calibration and GIS alignment challenges, and a long tail of edge cases that emerge in real-world operations.

This talk introduces Siemens Energy’s end-to-end overhead line inspection solution and shares key learnings from inspecting more than 10,000 km of power lines for real customers across several continents. We will show how raw 2D/3D data is transformed into structured information, delivering insights into asset inventory as well as defects, and supporting maintenance and planning decisions for critical infrastructure. The focus is on the combination of algorithmic building blocks and scalable processing, designed for robustness and consistency at scale, where even low error rates can become operationally significant.

Attending the 3D Vision Conference means confronting 177 accepted papers across 3.5 days, far more than any one person can absorb. Skimming titles the night before isn't enough.

This talk shows how to build a systematic, interactive map of an entire conference using modern open-source tools. We load all 177 papers from 3DV 2026 (full PDF page images plus metadata) into a FiftyOne grouped dataset. We then run three annotation passes using Qwen3.5-9B on each cover page: topic classification, author affiliation extraction, and project page detection. Document embeddings from Jina v4 are computed across all 3,019 page images, pooled to paper-level vectors, and fed into FiftyOne Brain for UMAP visualization, similarity search, representativeness scoring, and uniqueness scoring.

The result is an interactive dataset you can query, filter, and explore in the FiftyOne App. Sort by uniqueness to find distinctive work, filter by topic and sort by representativeness to understand each research area, and cross-reference with scheduling metadata to build a personal agenda.

I demonstrate the end-to-end pipeline and discuss design decisions regarding grouped datasets, reasoning model output parsing, and embedding pooling strategies.

AI agents are having a moment, but most of them are little more than fragile prototypes that break under pressure. Together, we’ll explore why so many agentic systems fail in practice, and how to fix that with real engineering principles. In this talk, you’ll learn how to build agents that are modular, observable, and ready for production. If you’re tired of shiny agent demos that don't deliver, this talk is your blueprint for building agents that actually work.

Drop in YouTube URLs or upload files and query content four ways: exact keyword matching, semantic search across transcripts, visual scene search via SigLIP2, and LLM-generated answers that synthesise across segments.