Research map / Detection and segmentation / CNN features
A Generalist Framework for Panoptic Segmentation of Images and Videos
Ting Chen, Lala Li, Saurabh Saxena, Geoffrey Hinton, David J. Fleet
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 |
|---|---|---|---|
| scale analog bits to 0.1 instead of 1.0 (ablation) | Pix2Seq-D · MS-COCO | PQ: 21.3 -> 43.9 | Improved |
| use softmax cross entropy loss instead of l2 loss (ablation) | Pix2Seq-D · MS-COCO | PQ: 41.9 -> 43.9 | Improved |
| add loss weighting for small instances (p=0.2) (ablation) | Pix2Seq-D · MS-COCO | PQ: 40.4 -> 43.9 | Improved |
| increase DDIM sampling steps from 5 to 20 (ablation) | Pix2Seq-D · MS-COCO val | PQ: 47.5 -> 50.3 | Improved |
| increase sampling steps from 20 to 50 (ablation) | Pix2Seq-D · MS-COCO val | PQ: 50.3 -> 50.2 | No clear effect |
| use larger image and mask size (ablation) | Pix2Seq-D · Cityscapes val | PQ: 52.1 -> 59.9 | Improved |
| condition on past mask predictions for video | Pix2Seq-D · DAVIS 2017 val | J&F: 68.4 | Improved |
| use 1 inference step for subsequent video frames (8 for first) (ablation) | Pix2Seq-D · DAVIS 2017 val | J&F: 67.3 vs 68.4 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Mask DINO — Mask DINO: Towards a Unified Transformer-Based Framework for Object Detection and Segmentation (CVPR 2023) · better on 0 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
- A Simple Latent Diffusion Approach for Panoptic Segmentation and Mask Inpainting (ECCV 2024 · similarity 0.91)
- Panoptic Segmentation (CVPR 2019 · similarity 0.91)
- Video Panoptic Segmentation (CVPR 2020 · similarity 0.91)
- Combinatorial Optimization for Panoptic Segmentation: A Fully Differentiable Approach (NeurIPS 2021 · similarity 0.90)
- Slot-VPS: Object-Centric Representation Learning for Video Panoptic Segmentation (CVPR 2022 · similarity 0.90)
- Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 · similarity 0.90)
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.
