Towards Perspective-Free Object Counting with Deep Learning

By Daniel Oñoro-Rubio, Roberto Javier López-Sastre. 2016

638 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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

Citations in the corpus

DeepSetNet: Predicting Sets with Deep Neural Networks (2016, methodology): “We follow exactly the same data split used in [32] by conducting four different and separate experiments on maximal, downscale, upscale and minimal subsets in UCSD dataset.”
Domain Adaptation for Traffic Density Estimation (2021, methodology): “A similar idea is developed by (Oñoro-Rubio and López-Sastre, 2016) with a scale-aware, multicolumn counting model named Hydra-CNN able to estimate traffic densities in congested scenes.”
DecideNet: Counting Varying Density Crowds Through Attention Guided Detection and Density Estimation (2017, methodology): “Traditional crowd counting by density estimation methods regress density maps by minimizing the pixel-wise Euclidean loss to the ground-truth [36, 2, 42, 22].”
MoVie: Revisiting Modulated Convolutions for Visual Counting and Beyond (2021, background): “…for specialized objects has a number of practical applications (Marsden et al., 2018), including but are not limited to cell counting (Xie et al., 2018), crowd counting (Sindagi & Patel, 2017), vehicle counting (Onoro-Rubio & L´opez-Sastre, 2016), wild-life counting (Arteta et al., 2016), etc .”
Where are the Blobs: Counting by Localization with Point Supervision (2018, methodology): “Onoro-Rubio et al. [10] extended this method by proposing a perspective-free multi-scale deep learning approach.”
Counting Vehicles with Cameras (2018, background): “Some works that use CNNs are [4] and [3] for crowd counting, and [12] for the vehicles counting task.”
Crowd Counting and Density Estimation by Trellis Encoder-Decoder Networks (2019, methodology): “Moreover, we also outperform MCNN on all datasets in terms of MAE and MSE.”
Generating High-Quality Crowd Density Maps Using Contextual Pyramid CNNs (2017, methodology): “As discussed in Section 1, though recent state-of-the-art multi-column or multi-scale methods [50, 23, 36] achieve significant improvements in the task of crowd count estimation, they either underestimate or overestimate counts in high-density and low-density crowd images respectively (as explained ”
Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds (2018, methodology): “CNN based approaches for crowd counting include [16], [29], [30], [19], [2].”
Incorporating Side Information by Adaptive Convolution (2017, methodology): “Onoro-Rubio and López-Sastre (2016) uses a pyramid of input patches, each sent to separate sub-network, to consider multiple scales, while Zhang et al. (2016) uses a multi-column CNN with different receptive field sizes in order to encourage different columns to capture features at different scales.”
Deep Count: Fruit Counting Based on Deep Simulated Learning (2017, neutral): “mean were the”
Context-Aware Crowd Counting (2018, methodology): “Previous works in crowd density estimation use the mean absolute error (MAE) and the root mean squared error (RMSE) as evaluation metrics [42, 41, 26, 31, 40, 36].”
JHU-CROWD++: Large-Scale Crowd Counting Dataset and A Benchmark Method (2020, methodology): “Typically, most existing works ([4, 5, 6, 7, 22, 23, 24, 25, 26, 26, 30, 31]) have designed their networks by laying a strong emphasis on addressing large variations of scale in crowd images.”
A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation (2017, methodology): “…and counting, they however cover only traditional methods that use hand-crafted features and do not take into account the recent advancements driven primarily by CNN-based approaches [87, 39, 113, 11, 85, 97, 4, 98, 111, 107, 70, 88] and creation of new challenging crowd datasets [106, 107, 111].”
Understanding Traffic Density from Large-Scale Web Camera Data (2017, methodology): “To verify the efficacy of our methods, we also evaluate and compare the proposed methods with baselines on a public dataset, TRANCOS[25].”
Crowd Counting via Adversarial Cross-Scale Consistency Pursuit (2018, methodology): “Recent CNN-based works [20, 37, 25] utilize multi-path architectures to address the scale variation issue and have achieved good improvements in crowd density estimation.”
Bayesian Loss for Crowd Count Estimation With Point Supervision (2019, background): “ing estimating the number of vehicles in traffic congestion [29, 30, 56, 14, 28], counting the cells and bacteria from microscopic images [20, 42, 44, 45, 8], and animal crowd estimations for ecological survey [27, 1, 18], to name a few.”