AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Multi-Task Driven Feature Models for Thermal Infrared Tracking

Qiao Liu, Xin Li, Zhenyu He, Nana Fan, Di Yuan, Wei Liu, Yongsheng Liang

AAAI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 79 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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
Add the auxiliary classification branch (ablation)MMNet · VOT-TIR2015EAO: 0.282 -> 0.307Improved
Add the auxiliary classification branch (ablation)MMNet · VOT-TIR2017EAO: 0.254 -> 0.274Improved
Add the holistic correlation module to fine-grained matching (ablation)MMNet with Dis + Cls · VOT-TIR2015Acc: 0.51 -> 0.58Improved
Add the holistic correlation module to fine-grained matching (ablation)MMNet with Dis + Cls · VOT-TIR2017Acc: 0.52 -> 0.55Improved
Add the pixel-level correlation module to fine-grained matching (ablation)MMNet with Dis + Cls · VOT-TIR2015Acc: 0.51 -> 0.57Improved
Add the pixel-level correlation module to fine-grained matching (ablation)MMNet with Dis + Cls · VOT-TIR2017Acc: 0.52 -> 0.56Improved
Use both holistic and pixel-level correlation modules (ablation)MMNet · VOT-TIR2015 and VOT-TIR2017EAO: 0.322 -> 0.332; 0.296 -> 0.320Improved
Mix VID2015 grayscale and TIR training data (ablation)MMNet · VOT-TIR2015EAO: 0.332 -> 0.344Improved
Mix VID2015 grayscale and TIR training data (ablation)MMNet · PTB-TIRSuc: 0.502 -> 0.539; 0.519 -> 0.539Improved
Retrain the overall network on TIR after VID pretraining (ablation)MMNet · VOT-TIR2015 and PTB-TIREAO / Suc: 0.332 -> 0.300; 0.502 -> 0.521No clear effect
Fine-tune VID initialization on TIR while freezing early layers and fine-grained branch (ablation)MMNet · PTB-TIRSuc: 0.502 -> 0.525Improved
Use MMNet rather than CFNetTIR tracking · PTB-TIRSuc: +10% relative gainImproved

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