AI Paper Atlas

Research map / Detection and segmentation / CNN features

Target-Aware Invertible Encoder with Reconstruction Guidance for Infrared Small Target Detection

Shule Yan, Zetian Zhang, Xiao Ma, Zexuan Ji

CVPR 2026 · Detection and segmentation › CNN features · approach: Context for hard detection conditions

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
increase InvBlocks per stage from [1,1,1,1] to [2,2,2,2] (Srev=2) (ablation)InvDet · IRSTD-1KF1 (%): 83.18 -> 84.40Improved
increase InvBlocks per stage from [2,2,2,2] to [4,4,4,4] (ablation)InvDet · IRSTD-1KF1 (%): 84.40 -> 84.10No clear effect
increase InvBlocks per stage from [2,2,2,2] to [3,3,3,3] (ablation)InvDet · NUDT-SIRSTF1 (%): 86.17 -> 88.90Improved
increase InvBlocks per stage from [2,2,2,2] to [3,3,3,3] (ablation)InvDet · IRSTD-1KF1 (%): 84.40 -> 83.00Hurt
increase reversible depth Srev from 2 to 4 with [2,2,2,2] (ablation)InvDet · IRSTD-1KF1 (%): 84.40 -> 82.62Hurt
increase reversible depth Srev from 2 to 4 with [2,2,2,2] (ablation)InvDet · NUAA-SIRSTF1 (%): 87.37 -> 89.29Improved
disable inverse reconstruction path at inference (ablation)InvDet · inference speedforward FPS: 50.30 E2E FPS -> 72.72 FWD FPSImproved
InvDet vs MA-Net on larger-scale datasetsInvDet · IRSTDF1 (%): 98.3 -> 97.8Hurt
InvDet vs second-best methodInvDet · IRSTD-1KF1 (%): 80.3 -> 84.4Improved
train on one infrared dataset and test on another without fine-tuningInvDet · cross-dataset F1mean retention (%): 84.9Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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