Camouflaged Object Detection

By Deng-Ping Fan, Ge-Peng Ji, Guolei Sun, Ming-Ming Cheng, Jianbing Shen, Ling Shao. Computer Vision and Pattern Recognition, 2020

720 citations recorded by the source corpus.

We present a comprehensive study on a new task named camouflaged object detection (COD), which aims to identify objects that are “seamlessly” embedded in their surroundings. The high intrinsic similarities between the target object and the background make COD far more challenging than the traditional object detection task. To address this issue, we elaborately collect a novel dataset, called COD10K, which comprises 10,000 images covering camouflaged objects in various natural scenes, over 78 object categories. All the images are densely annotated with category, bounding-box, object-/instance-level, and matting-level labels. This dataset could serve as a catalyst for progressing many vision tasks, e.g., localization, segmentation, and alpha-matting, etc. In addition, we develop a simple but effective framework for COD, termed Search Identification Network (SINet). Without any bells and whistles, SINet outperforms various state-of-the-art object detection baselines on all datasets tested, making it a robust, general framework that can help facilitate future research in COD. Finally, we conduct a large-scale COD study, evaluating 13 cutting-edge models, providing some interesting findings, and showing several potential applications. Our research offers the community an opportunity to explore more in this new field. The code will be available at https://github.com/DengPingFan/SINet/.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/e5fb1c378086d072104b14f4fb2df9cf1da4b08c

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

Uncertainty-aware Joint Salient Object and Camouflaged Object Detection (2021, methodology): “In general, we can observe improved performance, especially for the COD10K dataset [11], which verifies effectiveness of our similarity measure module.”
PraNet: Parallel Reverse Attention Network for Polyp Segmentation (2020, background): “further introduce four other metrics which are widely used in the field of object detection [7,31,8,11,37,38].”
Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection (2022, methodology): “There are also a number of bio-inspired methods, such as [7, 46].”
Kubric: A scalable dataset generator (2022, methodology): “Classical methods involve active-contour [47, 69] or graphcut [20, 104] techniques, but there also exist techniques with human-in-the-loop [72, 81, 100], and more recently deep learning variants [30, 73, 74, 105, 117].”
Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers (2021, background): “similar to surrounding tissues, with low contrast, providing them with powerful camouflage properties [11], [12], and making them difficult to identify; (3) Polycentric data.”
Concealed Object Detection (2021, methodology): “A preliminary version of this work has appeared in CVPR 2020 [1].”
Simultaneously Localize, Segment and Rank the Camouflaged Objects (2021, methodology): “The training set includes 1,711 images from the COD10K-CAM training set [10] and 289 images are from the CAMO training set [22].”
UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders (2020, methodology): “Furthermore, we plan to capture new datasets with multiple human annotations to further model the statistics of human uncertainty in interactive image segmentation [37], camouflaged object detection [16], etc.”
Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks (2019, background): “, PDC in [112]) or training strategy [113] are promising to increase the performance.”
Camouflaged Object Detection with Feature Decomposition and Edge Reconstruction (2023, methodology): “To tackle these challenges, most existing works aim to develop models that mimic the human visual system [6,28].”
RGB-D salient object detection: A survey (2020, background): “SOD plays a key role in a range of real-world applications, such as stereo matching [11], image understanding [12], co-saliency detection [13], action recognition [14], video detection and segmentation [15]–[18], semantic segmentation [19], [20], medical image segmentation [21]–[23], object tracking”
Camouflaged Object Segmentation with Distraction Mining (2021, methodology): “COD10K [12] is currently the largest benchmark dataset, which includes 5,066 camouflaged images (3,040 for training and 2,026 for testing) downloaded from multiple photography websites, covering 5 super-classes and 69 sub-classes.”
Interactive Image Segmentation With First Click Attention (2020, background): “Recent years, with the popularization of data-driven deep learning techniques, the demand for mask-level annotations has increased dramatically in some fields, such as salient object detection [4, 9, 14, 23], semantic segmentation [34], instance segmentation [21, 35], camouflaged object detection [1”
Inf-Net: Automatic COVID-19 Lung Infection Segmentation From CT Images (2020, methodology): “In the future, we plan to apply our Inf-Net to other related tasks, such as polyp segmentation [84], product defects detection, camouflaged animal detection [85].”
Uncertainty-Guided Transformer Reasoning for Camouflaged Object Detection (2021, methodology): “, fixations [36], boundaries [57], image-level labels [12, 29], for joint representation learning.”