The Pascal Visual Object Classes Challenge: A Retrospective

By Mark Everingham, S. M. Ali Eslami, Luc Van Gool, Christopher K. I. Williams, John M. Winn, Andrew Zisserman. 2014

6,745 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Evaluation & Benchmarks

Citations in the corpus

Object-Centric Learning with Slot Attention (2020, methodology): “[81] Mark Everingham, SM Ali Eslami, Luc Van Gool, Christopher KI Williams, John Winn, and Andrew Zisserman.”
S4L: Self-Supervised Semi-Supervised Learning (2019, background): “Modern computer vision systems demonstrate outstanding performance on a variety of challenging computer vision benchmarks, such as image recognition [32], object detection [20], semantic image segmentation [8], etc.”
Weight Standardization (2019, methodology): “…experiments, including image classiﬁcation on ImageNet dataset [35], object detection and instance segmentation on COCO dataset [22], video recognition on Something-SomethingV1 dataset [8], semantic image segmentation on PASCAL VOC [6], and point cloud classiﬁcation on ModelNet40 dataset [43].”
Learning Visual Features from Large Weakly Supervised Data (2015, methodology): “…a range of computer-vision tasks: (1) the MIT Indoor dataset [38], (2) the MIT SUN dataset [55], (3) the Stanford 40 Actions dataset [57], (4) the Oxford Flowers dataset [33], (5) the Sports dataset [17], (6) the ImageNet ILSVRC 2014 dataset [42], and (7) the Pascal VOC 2007 dataset [11].”
Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles (2016, methodology): “quantitatively on image retrieval. 4.1 Transfer Learning We evaluate our learned features as pre-trained weights for classication, detection, and semantic segmentation tasks on the PASCAL VOC dataset[12]. We also introduce a novel benchmark to evaluate methods for unsupervised/selfsupervised represe”
Panoptic Segmentation (2018, methodology): “The PASCAL VOC dataset [9] upgraded the task to a more diverse set of general object classes on more challenging images.”
End-to-End People Detection in Crowded Scenes (2015, methodology): “We conduct evaluation using the standard protocol defined in [4].”
You Only Look Once: Unified, Real-Time Object Detection (2015, methodology): “We implement this model as a convolutional neural network and evaluate it on the PASCAL VOC detection dataset [9].”
The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, methodology): “As detection results have matured (70% mean AP on PASCAL [14, 55]), the last years have seen a rising interest in more difficult settings.”
What Makes for Effective Detection Proposals? (2015, methodology): “We test the generalisation of proposal methods by evaluating on datasets with many additional classes in §4.”
COCO-Stuff: Thing and Stuff Classes in Context (2016, background): “These datasets have bounding box or outline-level annotations of things, e.g. PASCAL VOC [18], ILSVRC [43], COCO [35].”
Self-Supervised Learning of Pretext-Invariant Representations (2019, methodology): “The performance of PIRL is within 2% of fully supervised representations on Places205, and improves the previous best results of [21] on VOC07 by more than 16 AP points.”
Scaling and Benchmarking Self-Supervised Visual Representation Learning (2019, methodology): “Without any bells and whistles, our self-supervised model initialization matches the performance of the supervised initialization on both VOC07 and VOC07+12.”
DensePose: Dense Human Pose Estimation in the Wild (2018, background): “Gathering rich, high-quality training sets has been a catalyst for progress in the classification [40], detection and segmentation [8, 27] tasks.”
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (2016, background): “The PASCAL VOC 2012 segmentation benchmark [34] involves 20 foreground object classes and one background class.”
Adversarial Feature Learning (2016, methodology): “9 Table 3: Classification and detection results for the PASCAL VOC 2007 [6] test set, and segmentation results on the PASCAL VOC 2012 [6] validation set, under the standard mean average precision (mAP) or mean intersection over union (mIU) metrics for each task.”
Language-driven Semantic Segmentation (2022, methodology): “The number of class labels is dictated by the training dataset and typically ranges from tens (Everingham et al., 2015) to hundreds (Zhou et al., 2019; Mottaghi et al., 2014) of distinct categories.”