AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Neural Baby Talk

Jiasen Lu, Jianwei Yang, Dhruv Batra, Devi Parikh

CVPR 2018 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 450 citations (OpenAlex)

Read the paper

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
Generate a hybrid caption template with region-grounded slots, then fill slots using detector conceptsImage captioning · Flickr30k Entities testCIDEr: 53.1 -> 57.5Improved
Generate a hybrid caption template with region-grounded slots, then fill slots using detector conceptsImage captioning · COCO testBLEU4: 33.4 -> 34.7Improved
Generate a hybrid caption template with region-grounded slots, then fill slots using detector conceptsImage captioning · COCO testCIDEr: 120.1 -> 107.2Hurt
Use ground-truth object regions and categories at test time (ablation)NBT · Robust COCO testAccuracy: 42.4 -> 45.7Improved
Use ground-truth object regions and categories at test time (ablation)NBT · COCO testBLEU4: 34.7 -> 34.9No clear effect
Use constrained beam search with 2 top detected concepts instead of greedy decoding (ablation)NBT with ResNet-101 · Novel object captioning out-of-domain testAverage F1: 53.2 -> 70.3Improved
Use constrained beam search with 1 top detected concept instead of greedy decoding (ablation)NBT with ResNet-101 · Novel object captioning out-of-domain testAverage F1: 53.2 -> 57.3Improved
Use 2 rather than 1 top detected concept in constrained beam search (ablation)NBT with ResNet-101 · Novel object captioning out-of-domain testAverage F1: 57.3 -> 70.3Improved
Use ResNet-101 instead of VGG-16 with greedy decoding (ablation)NBT · Novel object captioning out-of-domain testAverage F1: 48.5 -> 53.2Improved

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