{"url":"/method/panoptic-fpn","slug":"panoptic-fpn","name":"Panoptic FPN","full_name":"Panoptic FPN","full_name_withheld":false,"description_markdown":"A **Panoptic FPN** is an extension of an [FPN](https://paperswithcode.com/method/fpn) that can generate both instance and semantic segmentations via FPN. The approach starts with an FPN backbone and adds a branch for performing semantic segmentation in parallel with the existing region-based branch for instance segmentation. No changes are made to the FPN backbone when adding the dense-prediction branch, making it compatible with existing instance segmentation methods. \r\n\r\nThe new semantic segmentation branch achieves its goal as follows. Starting from the deepest FPN level (at 1/32 scale), we perform three upsampling stages to yield a feature map at 1/4 scale, where each upsampling stage consists of 3×3 [convolution](https://paperswithcode.com/method/convolution), group norm, [ReLU](https://paperswithcode.com/method/relu), and 2× bilinear upsampling. This strategy is repeated for FPN scales 1/16, 1/8, and 1/4 (with progressively fewer upsampling stages). The result is a set of feature maps at the same 1/4 scale, which are then element-wise summed. A final 1×1 convolution, 4× bilinear upsampling, and [softmax](https://paperswithcode.com/method/softmax) are used to generate the per-pixel class labels at the original image resolution. In addition to stuff classes, this branch also outputs a special ‘other’ class for all pixels belonging to objects (to avoid predicting stuff classes for such pixels).","description_state":"present","introduced_year":null,"introduced_by":{"title":"Panoptic Feature Pyramid Networks","paper":"/paper/panoptic-feature-pyramid-networks","first_author":"Alexander Kirillov","n_authors":4,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/panoptic-feature-pyramid-networks"},"source":{"url":"http://arxiv.org/abs/1901.02446v2","title":"Panoptic Feature Pyramid Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/facebookresearch/detectron2/blob/1b09e42cc87d47a6e0a3892cd86e780d86a9b122/detectron2/modeling/meta_arch/panoptic_fpn.py#L20","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Feature Extractors","url":"/methods/category/feature-extractors","pwc_aliases":[]}],"n_papers_tagged":3,"archive_num_papers":3,"papers_newest_first":[{"paper":"/paper/pointrend-image-segmentation-as-rendering","title":"PointRend: Image Segmentation as Rendering","date":"2019-12-17","arxiv_id":"1912.08193","n_code_links":14,"syntology":{"ran":5,"of":18,"unverified":13,"pointer_only":0}},{"paper":null,"title":"IMP: Instance Mask Projection for High Accuracy Semantic Segmentation of Things","date":"2019-06-15","arxiv_id":"1906.06597","n_code_links":0,"syntology":null},{"paper":"/paper/panoptic-feature-pyramid-networks","title":"Panoptic Feature Pyramid Networks","date":"2019-01-08","arxiv_id":"1901.02446","n_code_links":12,"syntology":{"ran":7,"of":12,"unverified":5,"pointer_only":0}}],"papers_shown":3,"tasks":[{"task":"/task/instance-segmentation","name":"Instance Segmentation","papers":3},{"task":"/task/segmentation","name":"Segmentation","papers":3},{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":3},{"task":"/task/image-segmentation","name":"Image Segmentation","papers":1},{"task":"/task/panoptic-segmentation","name":"Panoptic Segmentation","papers":1},{"task":"/task/scene-segmentation","name":"Scene Segmentation","papers":1},{"task":"/task/thermal-image-segmentation","name":"Thermal Image Segmentation","papers":1},{"task":"/task/high","name":"Vocal Bursts Intensity Prediction","papers":1}],"tasks_shown":8,"n_tasks":8,"usage_by_year":[{"year":"2019","papers":3}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/panoptic-fpn"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}