Papers › Mask Frozen-DETR: High Quality Instance Segmentation with One GPU

Mask Frozen-DETR: High Quality Instance Segmentation with One GPU

7 Aug 2023arXiv:2308.03747archive 2025-07-28

Zhanhao Liang, Yuhui Yuan

In this paper, we aim to study how to build a strong instance segmenter with minimal training time and GPUs, as opposed to the majority of current approaches that pursue more accurate instance segmenter by building more advanced frameworks at the cost of longer training time and higher GPU requirements. To achieve this, we introduce a simple and general framework, termed Mask Frozen-DETR, which can convert any existing DETR-based object detection model into a powerful instance segmentation model. Our method only requires training an additional lightweight mask network that predicts instance masks within the bounding boxes given by a frozen DETR-based object detector. Remarkably, our method outperforms the state-of-the-art instance segmentation method Mask DINO in terms of performance on the COCO test-dev split (55.3% vs. 54.7%) while being over 10X times faster to train. Furthermore, all of our experiments can be trained using only one Tesla V100 GPU with 16 GB of memory, demonstrating the significant efficiency of our proposed framework.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Instance SegmentationObject DetectionSemantic Segmentationobject-detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Instance Segmentation COCO minival Mask Frozen-DETR AP50 78.9 #5 of 93 Archive leaderboard report
Instance Segmentation COCO minival Mask Frozen-DETR AP75 60.8 #5 of 93 Archive leaderboard report
Instance Segmentation COCO minival Mask Frozen-DETR APL 72.9 #5 of 93 Archive leaderboard report
Instance Segmentation COCO minival Mask Frozen-DETR APM 58.4 #5 of 93 Archive leaderboard report
Instance Segmentation COCO minival Mask Frozen-DETR APS 37.2 #5 of 93 Archive leaderboard report
Instance Segmentation COCO minival Mask Frozen-DETR mask AP 54.9 #5 of 93 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR AP50 79.3 #5 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR AP75 61.4 #5 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR APL 70.4 #5 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR APM 58.4 #5 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR APS 37.8 #5 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev Mask Frozen-DETR mask AP 55.3 #5 of 112 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionDense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections