# Camera clustering for scalable stream-based active distillation

Source: [Camera clustering for scalable stream-based active distillation](<https://www.sota2.com/research/paper/68fa72f21e209131cea597c0>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:38:38.445Z

Authors: Dani Manjah, Davide Cacciarelli, Christophe De Vleeschouwer, Benoit Macq

Published: 2024-04-16T09:28:54.000Z

## Abstract

We present a scalable framework designed to craft efficient lightweight models for video object detection utilizing self-training and knowledge distillation techniques. We scrutinize methodologies for the ideal selection of training images from video streams and the efficacy of model sharing across numerous cameras. By advocating for a camera clustering methodology, we aim to diminish the requisite number of models for training while augmenting the distillation dataset. The findings affirm that proper camera clustering notably amplifies the accuracy of distilled models, eclipsing the methodologies that employ distinct models for each camera or a universal model trained on the aggregate camera data.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2404.10411>)

[PDF](<https://arxiv.org/pdf/2404.10411.pdf>)
