Research map / Detection and segmentation / Weakly supervised detection and segmentation
PPR-FCN: Weakly Supervised Visual Relation Detection via Parallel Pairwise R-FCN
Hanwang Zhang, Zawlin Kyaw, Jinyang Yu, Shih-Fu Chang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add role (subject/object) sensitivity to the score map (ablation) | PPR-FCN predicate prediction · VRD predicate prediction | R@50: 24.30 -> 42.74 | Improved |
| add role sensitivity to the score map (ablation) | PPR-FCN predicate prediction · VG predicate prediction | R@50: 43.30 -> 61.57 | Improved |
| use pairwise RoI pooling with role-sensitive maps instead of position-sensitive map with joint box pooling (ablation) | PPR-FCN predicate prediction · VRD predicate prediction | R@50: 29.57 -> 42.74 | Improved |
| replace pooled scores with fully-connected subnetworks (ablation) | PosSeq+Pairwise · VRD predicate prediction | R@50: 42.74 -> 47.43 | Improved |
| replace pooled scores with fully-connected subnetworks (ablation) | PosSeq+Pairwise · VG predicate prediction | R@50: 61.57 -> 64.17 | Improved |
| add parallel pair-selection branch to classification branch (ablation) | PPR-FCN · VRD phrase detection | R@50: 3.56 -> 6.93 | Improved |
| add parallel pair-selection branch to classification branch (ablation) | PPR-FCN · VG phrase detection | R@50: 0.87 -> 2.41 | Improved |
| add parallel pair-selection branch to classification branch (ablation) | PPR-FCN · VRD relation detection | R@50: 3.56 -> 5.68 | Improved |
| use PPR-FCN spatial-context pairwise pooling instead of VTransE-MIL | weakly supervised relation detection · VRD relation detection | R@50: 4.28 -> 5.68 | Improved |
| use the parallel FCN WSOD module instead of WSDDN | weakly supervised object detection · VRD object detection | R@50: 15.08 -> 25.34 | Improved |
| use original EdgeBox scores instead of the pre-trained objectness-filtered top 1,000 proposals (ablation) | WSOD module · VRD/VG object detection | R@K: drop by about 5 points | Hurt |
| use fully convolutional pairwise pooling instead of per-region fc subnetworks | PPR-FCN · test time | ms/img: 270 -> 150 | Improved |
Later papers that compared with this method
4 later accepted papers report PPR-FCN as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- Weakly-supervised HOI Detection via Prior-guided Bi-level Representation Learning (ICLR 2023 · better on 1 of 1 shared result)
- Weakly Supervised Visual Semantic Parsing (CVPR 2020 · better on 4 of 4 shared results)
- On Exploring Undetermined Relationships for Visual Relationship Detection (CVPR 2019 · better on 1 of 1 shared result)
- Zoom-Net: Mining Deep Feature Interactions for Visual Relationship Recognition (ECCV 2018 · better on 1 of 1 shared result)
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
- Localize, Assemble, and Predicate: Contextual Object Proposal Embedding for Visual Relation Detection (AAAI 2020 · similarity 0.91)
- Explanation-based Weakly-supervised Learning of Visual Relations with Graph Networks (ECCV 2020 · similarity 0.91)
- MR-NET: Exploiting Mutual Relation for Visual Relationship Detection (AAAI 2019 · similarity 0.90)
- Weakly-Supervised Learning of Visual Relations (ICCV 2017 · similarity 0.89)
- One-Shot Learning for Long-Tail Visual Relation Detection (AAAI 2020 · similarity 0.89)
- RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder (NeurIPS 2020 · similarity 0.88)
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.
