AI Paper Atlas

Research map / Detection and segmentation / CNN features

Visualizing Object Detection Features

Carl Vondrick, Aditya Khosla, Hamed Pirsiavash, Tomasz Malisiewicz, Antonio B. Torralba

IJCV 2016 · Detection and segmentation › CNN features · approach: Box regression and loss design · 41 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
use paired dictionary learning instead of the standard HOG glyphHOG visualization · MTurk classification of PASCAL VOC windowsaccuracy: 0.191 -> 0.383Improved
use direct optimization instead of the HOG glyphHOG visualization · MTurk classification of PASCAL VOC windowsaccuracy: 0.191 -> 0.355Improved
use exemplar LDA instead of paired dictionaryHOG inversion · Pixel-level reconstruction (NCC)mean NCC: 0.637 -> 0.671Improved
use exemplar LDA instead of paired dictionaryHOG inversion · MTurk classification of inversionsaccuracy: 0.383 -> 0.282Hurt
use ridge regression instead of exemplar LDAHOG inversion · Pixel-level reconstruction (NCC)mean NCC: 0.671 -> 0.656Hurt
use expert PhD students instead of MTurk workers on glyphsHOG glyph · MTurk-style classificationaccuracy: 0.191 -> 0.233Improved
experts viewing paired dictionary instead of non-expertsHOG visualization · classificationaccuracyNo clear effect
use glyph for bicycles instead of paired dictionaryHOG visualization · MTurk classification, bicycleaccuracy: 0.307 -> 0.405Improved
use RGB patches instead of HOG visualization for human chair detectionHuman sliding-window detector · PASCAL VOC chair detectionaccuracyImproved
add color-averaging affinity diversity cost to multiple inversions (ablation)paired dictionary on CNN pool5 features · Multiple inversions, VOC 2007 windowsimage distance at fixed feature distanceImproved
add edge affinity diversity cost to multiple inversions (ablation)paired dictionary on CNN pool5 features · Multiple inversions, VOC 2007 windowsimage distance at fixed feature distanceImproved
use nudged or subset dictionaries instead of diversity cost (ablation)multiple inversions baseline · Multiple inversions, VOC 2007 windowsdiversity vs feature matchHurt

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