Box2Pix: Single-Shot Instance Segmentation by Assigning Pixels to Object Boxes

By Jonas Uhrig, Eike Rehder, Björn Fröhlich, Uwe Franke, Thomas Brox. 2018

68 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation (2020, methodology): “Recently, PanopticDeepLab [20], a simple, fast, and strong approach for bottom-up panoptic segmentation, employs a class-agnostic instance segmentation branch involving a simple instance center regression [45,82,66], coupled with DeepLab semantic segmentation outputs [13,15,16].”
MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers (2020, methodology): “Then, instance segments are obtained by grouping ‘thing’ pixels with various methods, such as instance center regression [44, 86, 70, 100, 20], Watershed transform [88, 3, 8], Hough-voting [4, 53, 8], or pixel affinity [45, 66, 81, 30, 8].”
YOLACT: Real-Time Instance Segmentation (2019, background): “Straight to Shapes [21] and Box2Pix [42] can perform instance segmentation in real-time (30 fps on Pascal SBD 2012 [12, 16] for Straight to Shapes, and 10.9 fps on Cityscapes [5] and 35 fps on KITTI [15] for Box2Pix), but their accuracies are far from that of modern baselines.”
Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth (2019, methodology): “[25], where they first predict bounding boxes based on a single-shot detection method, and then associate pixels by pointing to object centers, which can afterwards be efficiently clustered.”
k-means Mask Transformer (2022, methodology): “…clustering-based methods usually operate on semantic segments [13,15,18] and group ‘thing’ pixels into instance segments with various representations, such as instance center regression [50,89,77,104,22,94,63], Watershed transform [91,5], Hough-voting [6,59,92], or pixel aﬃnity [51,69,85,36,47].”
CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation (2022, methodology): “Other proxy-based methods typically start with semantic segments [11,13,16] and group ‘thing’ pixels into instance segments with various proxy tasks, such as instance center regression [19,42,56,67,79,83,90], Watershed transform [4, 8, 81], Hough-voting [5, 8, 51], or pixel affinity [8, 29, 43, 61, ”
Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges (2019, background): “For example, [54], [60], [61] focus on pixel-wise semantic segmentation for multiple classes including road, car, bicycle, column-pole, tree, sky, etc; [52] and [62] concentrate on road segmentation; and [51], [63], [64] deal”
Scaling Wide Residual Networks for Panoptic Segmentation (2020, methodology): “Recently, Panoptic-DeepLab [12], a simple yet effective bottom-up system for panoptic segmentation, employs DeepLab semantic segmentation outputs [6, 8] coupled with a class-agnostic instance segmentation branch involving a simple instance center regression [41, 86, 64].”
Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation (2021, methodology): “The bottom-up methods [9, 36] widely employ multiple stages to first perform semantic segmentation and then identify the specific location of each instance by boundary detection [20], pixel clustering [25], or pixels embedding loss [30, 22, 31], position-sensitive pooling [11, 24].”
YOLACT++ Better Real-Time Instance Segmentation (2019, background): “Straight to Shapes [34] and Box2Pix [35] can perform instance segmentation in realtime (30 fps on Pascal SBD 2012 [36], [37] for Straight to Shapes, and 10.9 fps on Cityscapes [38] and 35 fps on KITTI [39] for Box2Pix), but their accuracies are far from that of modern baselines.”
Retina U-Net: Embarrassingly Simple Exploitation of Segmentation Supervision for Medical Object Detection (2018, background): “of a single-shot detection (SSD) architecture for instance segmentation, where segmentation outputs are assigned to box proposals in a post-processing step [25].”