AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Towards Latent Attribute Discovery From Triplet Similarities

Ishan Nigam, Pavel Tokmakov, Deva Ramanan

ICCV 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models · 10 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 end-to-end training of the embedding to MVTE (compare LSN with end-to-end MVTE)LSN · Zappos-HumanSupervised-Eval accuracy: end-to-end MVTE 3.14% worse than LSNImproved
compare LSN with end-to-end MVTE on Celeb-ALSN · Celeb-A FacesSupervised-Eval accuracy: end-to-end MVTE 4.72% worse than LSNImproved
use hard assignment of triplets to latent embeddings instead of soft assignment (MVTE)LSN · UT-Zappos Zappos-HumanSupervised-Eval accuracy: 77.94 -> 88.91Improved
learn multiple latent attribute embeddings instead of a single embedding (SSN)LSN · Celeb-A Faces 8 attributesSupervised-Eval accuracy: 71.40 -> 83.53Improved
add attribute supervision (CSN vs LSN)LSN · Zappos-HumanSupervised-Eval accuracy: 88.91 -> 97.36Improved
increase number of underlying attributes from 2 to 8LSN · Celeb-A FacesSupervised-Eval accuracy: 92.33 -> 83.53Hurt
use 4 latent embeddings instead of 2 on Zappos-Human (ablation)LSN · Zappos-HumanSupervised-Eval accuracyNo clear effect
use 4 latent embeddings instead of 2 on Zappos-Meta (ablation)LSN · Zappos-MetaSupervised-Eval accuracyImproved
replace learned attribute clustering with random assignment of 4 attributes to 2 embeddings (ablation)LSN · Zappos-HumanSupervised-Eval accuracy: 6.69±0.53% lower with random assignmentImproved
replace learned attribute clustering with random assignment of 40 attributes to 8 embeddings (ablation)LSN · Celeb-A Faces (40 attributes)Supervised-Eval accuracy: 3.21 ± 0.68% lower with random assignmentImproved
increase latent embeddings beyond 8 on Celeb-A (40 attributes) (ablation)LSN · Celeb-A Faces (40 attributes)Supervised-Eval accuracyHurt

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