atlas

Research map / Detection and segmentation / Tracking

Siamese Instance Search for Tracking

CVPR 2016 · Detection and segmentation › Tracking · approach: Transformer-based video object tracking · 1,156 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Fine-tune the Siamese matching network on external ALOV videos rather than using ImageNet features aloneAlexNet-style Siamese tracker · OTBAUC: 42.8 -> 47.4Helped
Fine-tune on the target's first frame instead of external video dataAlexNet-style Siamese tracker · OTBAUC: 42.8 -> 44.0No clear effect
Remove max pooling (ablation)AlexNet tracker using pretrained fc6 features · OTBAUC: 42.8 -> 50.0Helped
Remove max pooling (ablation)ALOV-finetuned AlexNet Siamese tracker · OTBAUC: 47.4 -> 53.9Helped
Use multi-layer conv4, conv5, and fc6 features rather than fc6 alone (ablation)ALOV-finetuned AlexNet Siamese tracker without max pooling · OTBAUC: 53.9 -> 55.0Helped
Use a deeper VGG16-style network rather than AlexNet (ablation)Siamese tracker with multi-layer features and no max pooling · OTBAUC: 55.0 -> 59.2Helped
Use adaptive candidate sampling and optical-flow filteringSINT · OTBAUCHelped
Use a fixed initial target template and a generic learned matching function without online updatingSINT · OTBAUCHelped

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy