Concealed Object Detection

20 Feb 2021  ·  Deng-Ping Fan, Ge-Peng Ji, Ming-Ming Cheng, Ling Shao ·

We present the first systematic study on concealed object detection (COD), which aims to identify objects that are "perfectly" embedded in their background. The high intrinsic similarities between the concealed objects and their background make COD far more challenging than traditional object detection/segmentation. To better understand this task, we collect a large-scale dataset, called COD10K, which consists of 10,000 images covering concealed objects in diverse real-world scenarios from 78 object categories. Further, we provide rich annotations including object categories, object boundaries, challenging attributes, object-level labels, and instance-level annotations. Our COD10K is the largest COD dataset to date, with the richest annotations, which enables comprehensive concealed object understanding and can even be used to help progress several other vision tasks, such as detection, segmentation, classification, etc. Motivated by how animals hunt in the wild, we also design a simple but strong baseline for COD, termed the Search Identification Network (SINet). Without any bells and whistles, SINet outperforms 12 cutting-edge baselines on all datasets tested, making them robust, general architectures that could serve as catalysts for future research in COD. Finally, we provide some interesting findings and highlight several potential applications and future directions. To spark research in this new field, our code, dataset, and online demo are available on our project page: http://mmcheng.net/cod.

PDF Abstract

Results from the Paper


 Ranked #1 on Camouflaged Object Segmentation on CAMO (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Camouflaged Object Segmentation CAMO SINet-V2 MAE 0.070 # 1
Weighted F-Measure 74.3 # 1
S-Measure 82.0 # 1
E-Measure 88.2 # 1
Camouflaged Object Segmentation COD SINet-V2 MAE 0.037 # 1
Weighted F-Measure 68.0 # 1
S-Measure 81.5 # 1
E-Measure 88.7 # 1
Dichotomous Image Segmentation DIS-TE1 SINetV2 max F-Measure 0.644 # 12
weighted F-measure 0.558 # 10
MAE 0.094 # 11
S-Measure 0.727 # 9
E-measure 0.791 # 10
HCE 274 # 17
Dichotomous Image Segmentation DIS-TE2 SINetV2 max F-Measure 0.700 # 16
weighted F-measure 0.618 # 14
MAE 0.099 # 14
S-Measure 0.753 # 15
E-measure 0.823 # 13
HCE 593 # 17
Dichotomous Image Segmentation DIS-TE3 SINetV2 max F-Measure 0.730 # 15
weighted F-measure 0.641 # 15
MAE 0.096 # 14
S-Measure 0.766 # 14
E-measure 0.849 # 12
HCE 1096 # 16
Dichotomous Image Segmentation DIS-TE4 SINetV2 max F-Measure 0.699 # 18
weighted F-measure 0.616 # 17
MAE 0.113 # 17
S-Measure 0.744 # 16
E-measure 0.824 # 13
HCE 3683 # 10
Dichotomous Image Segmentation DIS-VD SINetV2 max F-Measure 0.665 # 16
weighted F-measure 0.584 # 15
MAE 0.110 # 14
S-Measure 0.727 # 16
E-measure 0.798 # 14
HCE 1568 # 14

Methods


No methods listed for this paper. Add relevant methods here