Rank & Sort Loss for Object Detection and Instance Segmentation

Date

2021-10-17

Author

ÖKSÜZ, KEMAL
ÇAM, BARIŞ CAN
Akbaş, Emre
Kalkan, Sinan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

129
views

0
downloads

We propose Rank & Sort (RS) Loss, a ranking-based loss function to train deep object detection and instance segmentation methods (i.e. visual detectors). RS Loss supervises the classifier, a sub-network of these methods, to rank each positive above all negatives as well as to sort positives among themselves with respect to (wrt.) their localisation qualities (e.g. Intersection-over-Union - IoU). To tackle the non-differentiable nature of ranking and sorting, we reformulate the incorporation of error-driven update with backpropagation as Identity Update, which enables us to model our novel sorting error among positives. With RS Loss, we significantly simplify training: (i) Thanks to our sorting objective, the positives are prioritized by the classifier without an additional auxiliary head (e.g. for centerness, IoU, mask-IoU), (ii) due to its ranking-based nature, RS Loss is robust to class imbalance, and thus, no sampling heuristic is required, and (iii) we address the multi-task nature of visual detectors using tuning-free task-balancing coefficients. Using RS Loss, we train seven diverse visual detectors only by tuning the learning rate, and show that it consistently outperforms baselines: e.g. our RS Loss improves (i) Faster R-CNN by 3 box AP and aLRP Loss (ranking-based baseline) by 2 box AP on COCO dataset, (ii) Mask R-CNN with repeat factor sampling (RFS) by 3.5 mask AP ( 7 AP for rare classes) on LVIS dataset; and also outperforms all counterparts. Code is available at: https://github.com/kemaloksuz/RankSortLoss.

URI

https://openaccess.thecvf.com/content/ICCV2021/html/Oksuz_Rank__Sort_Loss_for_Object_Detection_and_Instance_Segmentation_ICCV_2021_paper.html
https://hdl.handle.net/11511/95974

Conference Name

IEEE/CVF International Conference on Computer Vision (ICCV)

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Rank & Sort Loss for Object Detection and Instance Segmentation Oksuz, Kemal; Cam, Baris Can; Akbaş, Emre; Kalkan, Sinan (2021-01-01) We propose Rank & Sort (RS) Loss, a ranking-based loss function to train deep object detection and instance segmentation methods (i.e. visual detectors). RS Loss supervises the classifier, a sub-network of these methods, to rank each positive above all negatives as well as to sort positives among themselves with respect to (wrt.) their localisation qualities (e.g. Intersection-over-Union - IoU). To tackle the non-differentiable nature of ranking and sorting, we reformulate the incorporation of error-driven ...
A Ranking-based, Balanced Loss Function Unifying Classification and Localisation in Object Detection Öksüz, Kemal; Çam, Barış Can; Akbaş, Emre; Kalkan, Sinan (2020-12-06) We propose average Localisation-Recall-Precision (aLRP), a unified, bounded, balanced and ranking-based loss function for both classification and localisation tasks in object detection. aLRP extends the Localisation-Recall-Precision (LRP) performance metric (Oksuz et al., 2018) inspired from how Average Precision (AP) Loss extends precision to a ranking-based loss function for classification (Chen et al., 2020). aLRP has the following distinct advantages: (i) aLRP is the first ranking-based loss function fo...
Mask-aware IoU for Anchor Assignment in Real-time Instance Segmentation ÖKSÜZ, KEMAL; ÇAM, BARIŞ CAN; Kahraman, Fehmi; Baltacı, Zeynep Sonat; Kalkan, Sinan; Akbaş, Emre (2021-11-29) This paper presents Mask-aware Intersection-over-Union (maIoU) for assigning anchor boxes as positives and negatives during training of instance segmentation methods. Unlike conventional IoU or its variants, which only considers the proximity of two boxes; maIoU consistently measures the proximity of an anchor box with not only a ground truth box but also its associated ground truth mask. Thus, additionally considering the mask, which, in fact, represents the shape of the object, maIoU enables a more accura...
Integer Linear Programming Solution for the Multiple Query Optimization Problem Dokeroglu, Tansel; Bayir, Murat Ali; Coşar, Ahmet (2014-10-28) Multiple Query Optimization (MQO) is a technique for processing a batch of queries in such a way that shared tasks in these queries are executed only once, resulting in significant savings in the total evaluation. The first phase of MQO requires producing alternative query execution plans so that the shared tasks between queries are identified and maximized. The second phase of MQO is an optimization problem where the goal is selecting exactly one of the alternative plans for each query to minimize the tota...
Segmentation Driven Object Detection with Fisher Vectors Cinbiş, Ramazan Gökberk; Schmid, Cordelia (2013-01-01) We present an object detection system based on the Fisher vector (FV) image representation computed over SIFT and color descriptors. For computational and storage efficiency, we use a recent segmentation-based method to generate class-independent object detection hypotheses, in combination with data compression techniques. Our main contribution is a method to produce tentative object segmentation masks to suppress background clutter in the features. Re-weighting the local image features based on these masks...

Citation Formats

K. ÖKSÜZ, B. C. ÇAM, E. Akbaş, and S. Kalkan, “Rank & Sort Loss for Object Detection and Instance Segmentation,” presented at the IEEE/CVF International Conference on Computer Vision (ICCV), Amerika Birleşik Devletleri, 2021, Accessed: 00, 2022. [Online]. Available: https://openaccess.thecvf.com/content/ICCV2021/html/Oksuz_Rank__Sort_Loss_for_Object_Detection_and_Instance_Segmentation_ICCV_2021_paper.html.