Advances in Deep Concealed Scene Understanding

Abstract

Concealed scene understanding (CSU) is a hot computer vision topic aiming to perceive objects exhibiting camouflage. The current boom in terms of techniques and applications warrants an up-to-date survey. This can help researchers to better understand the global CSU field, including both current achievements and remaining challenges. This paper makes four contributions: (1) For the first time, we present a comprehensive survey of deep learning techniques aimed at CSU, including a taxonomy, task-specific challenges, and ongoing developments. (2) To allow for an authoritative quantification of the state-of-the-art, we offer the largest and latest benchmark for concealed object segmentation (COS). (3) To evaluate the generalizability of deep CSU in practical scenarios, we collect the largest concealed defect segmentation dataset termed CDS2K with the hard cases from diversified industrial scenarios, on which we construct a comprehensive benchmark. (4) We discuss open problems and potential research directions for CSU. Our code and datasets are available at https://github.com/DengPingFan/CSU, which will be updated continuously to watch and summarize the advancements in this rapidly evolving field.

arXiv comment18 pages, 6 figures, 8 tables

References in corpus (65)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Res2Net: A New Multi-scale Backbone Architecture
• PVT v2: Improved Baselines with Pyramid Vision Transformer
• Deep Visual Domain Adaptation: A Survey
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Salient Object Detection: A Benchmark
• Segmentation-Based Deep-Learning Approach for Surface-Defect Detection
• Concealed Object Detection
• Anabranch Network for Camouflaged Object Segmentation
• High-Resolution Image Synthesis with Latent Diffusion Models
• Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks
• Deeply supervised salient object detection with short connections
• Video Salient Object Detection via Fully Convolutional Networks
• Salient Object Detection: A Survey
• Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders
• Multi-interactive Dual-decoder for RGB-thermal Salient Object Detection
• Feature Aggregation and Propagation Network for Camouflaged Object Detection
• Fast Camouflaged Object Detection via Edge-based Reversible Re-calibration Network
• Re-thinking Co-Salient Object Detection
• RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
• MirrorNet: Bio-Inspired Camouflaged Object Segmentation
• Enhanced-alignment Measure for Binary Foreground Map Evaluation
• Exploring Rich and Efficient Spatial Temporal Interactions for Real Time Video Salient Object Detection
• Structure-measure: A New Way to Evaluate Foreground Maps
• Cascaded Partial Decoder for Fast and Accurate Salient Object Detection
• Video Salient Object Detection Using Spatiotemporal Deep Features
• SAM Struggles in Concealed Scenes -- Empirical Study on Segment Anything
• Light Field Salient Object Detection: A Review and Benchmark
• EGNet:Edge Guidance Network for Salient Object Detection
• Joint Learning of Saliency Detection and Weakly Supervised Semantic Segmentation
• Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective
• Camouflaged Object Detection and Tracking: A Survey
• Camouflaged Instance Segmentation In-The-Wild: Dataset, Method, and Benchmark Suite
• Salient Objects in Clutter
• Progressive Glass Segmentation
• Simultaneously Localize, Segment and Rank the Camouflaged Objects
• Salient Objects in Clutter: Bringing Salient Object Detection to the Foreground
• Regularized Densely-connected Pyramid Network for Salient Instance Segmentation
• Camouflaged Object Segmentation with Distraction Mining
• EM-driven unsupervised learning for efficient motion segmentation
• Context-aware Cross-level Fusion Network for Camouflaged Object Detection
• Uncertainty-aware Joint Salient Object and Camouflaged Object Detection
• Improving Camouflaged Object Detection with the Uncertainty of Pseudo-edge Labels
• Learning Uncertain Convolutional Features for Accurate Saliency Detection
• Weakly Supervised Salient Object Detection Using Image Labels
• Mutual Graph Learning for Camouflaged Object Detection
• Full-Duplex Strategy for Video Object Segmentation
• Selectivity or Invariance: Boundary-aware Salient Object Detection
• Masked Vision-Language Transformer in Fashion
• Learning Video Object Segmentation from Unlabeled Videos
• Motion Guided Attention for Video Salient Object Detection
• The Emergence of Objectness: Learning Zero-Shot Segmentation from Videos
• Segmenting Moving Objects via an Object-Centric Layered Representation
• High-resolution Iterative Feedback Network for Camouflaged Object Detection
• Weakly-Supervised Camouflaged Object Detection with Scribble Annotations
• PolarMask++: Enhanced Polar Representation for Single-Shot Instance Segmentation and Beyond
• CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification
• Feature Shrinkage Pyramid for Camouflaged Object Detection with Transformers
• Boosting Camouflaged Object Detection with Dual-Task Interactive Transformer
• Object Discovery in Videos as Foreground Motion Clustering
• Self-supervised Video Object Segmentation by Motion Grouping
• Dense Unsupervised Learning for Video Segmentation
• A Deeper Dive Into What Deep Spatiotemporal Networks Encode: Quantifying Static vs. Dynamic Information
• Locality-Aware Inter-and Intra-Video Reconstruction for Self-Supervised Correspondence Learning

Cited by in corpus (9)

• Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications
• ZoomNeXt: A Unified Collaborative Pyramid Network for Camouflaged Object Detection
• Bilateral Reference for High-Resolution Dichotomous Image Segmentation
• Rethinking Object Saliency Ranking: A Novel Whole-flow Processing Paradigm
• How Good is Google Bard's Visual Understanding? An Empirical Study on Open Challenges
• Rethinking Polyp Segmentation from an Out-of-Distribution Perspective
• Towards Real Zero-Shot Camouflaged Object Segmentation without Camouflaged Annotations
• Frontiers in Intelligent Colonoscopy
• Weakly Supervised Camouflaged Object Detection Based on the SAM Model and Mask Guidance