paper-with-me

홈 › Papers

An object-centric sensitivity analysis of deep learning based instance segmentation

2021-09-29 · Johannes Theodoridis, Jessica Hofmann, Johannes Maucher, Andreas Schilling

In this study we establish a comprehensive baseline regarding the object-centric robustness of deep learning models for instance segmentation. Our approach is motivated by the work of Geirhos et al. (2019) on texture bias in CNNs. However, we do not compare against human performance but instead incorporate ideas from object-centric representation learning. In addition, we analyze and control the effect of strong stylization that can lead to disappearing objects. The result is a stylized and object-centric version of MS COCO on which we perform an extensive sensitivity analysis regarding visual feature corruptions. We evaluate a broad range of frameworks including Cascade and Mask R-CNN, Swin Transformer, YOLACT(++), DETR, SOTR and SOLOv2. We find that framework choice, data augmentation and dynamic architectures improve robustness whereas supervised and self supervised pre-training does surprisingly not. In summary we evaluate 63 models on 61 versions of COCO for a total of 3843 evaluations.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Data AugmentationInstance SegmentationObjectRepresentation LearningSemantic SegmentationSensitivity

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
RPN A Region Proposal Network, or RPN, is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained…
Residual Connection 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Position-Wise Feed-Forward Layer 설명 없음

Similar Papers 제목 키워드 기반

3D-Aware Instance Segmentation and Tracking in Egocentric Videos

2024-08-19 · Yash Bhalgat, Vadim Tschernezki, Iro Laina, João F. Henriques 외

Egocentric videos present unique challenges for 3D scene understanding due to rapid camera motion, frequent object occlusions, and limited object visibility. This paper introduces a novel approach to instance segmentatio…

3D Object ReconstructionInstance SegmentationObjectObject Reconstruction+5

HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction

2022-03-03 · CVPR 2022 1 · Yunze Liu, Yun Liu, Che Jiang, Kangbo Lyu 외

We present HOI4D, a large-scale 4D egocentric dataset with rich annotations, to catalyze the research of category-level human-object interaction. HOI4D consists of 2.4M RGB-D egocentric video frames over 4000 sequences c…

Action SegmentationBenchmarkingHuman-Object Interaction DetectionMotion Segmentation+5

Sensitivity of Slot-Based Object-Centric Models to their Number of Slots

2023-05-30 · Roland S. Zimmermann, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Thomas Kipf 외

Self-supervised methods for learning object-centric representations have recently been applied successfully to various datasets. This progress is largely fueled by slot-based methods, whose ability to cluster visual scen…

Sensitivity

InstMove: Instance Motion for Object-centric Video Segmentation

2023-03-14 · CVPR 2023 1 · Qihao Liu, Junfeng Wu, Yi Jiang, Xiang Bai 외

Despite significant efforts, cutting-edge video segmentation methods still remain sensitive to occlusion and rapid movement, due to their reliance on the appearance of objects in the form of object embeddings, which are …

ObjectOptical Flow EstimationSegmentationVideo Segmentation+1

EgoLifter: Open-world 3D Segmentation for Egocentric Perception

2024-03-26 · Qiao Gu, Zhaoyang Lv, Duncan Frost, Simon Green 외

In this paper we present EgoLifter, a novel system that can automatically segment scenes captured from egocentric sensors into a complete decomposition of individual 3D objects. The system is specifically designed for eg…

3D ReconstructionObject