Publications by Simone Calderara

The slaughterhouse can act as a valid checkpoint to estimate the prevalence and the economic impact of diseases in farm animals. At present, scoring lesions is a challenging and time‐consuming activity, which is carried out by veterinarians serving the slaughter chain. Over recent years, artificial intelligence(AI) has gained traction in many fields of research, including livestock production. In particular, AI‐based methods appear able to solve highly repetitive tasks and to consistently analyze large amounts of data, such as those collected by veterinarians during postmortem inspection in high‐throughput slaughterhouses. The present study aims to develop an AI‐based method capable of recognizing and quantifying enzootic pneumonia‐like lesions on digital images captured from slaughtered pigs under routine abattoir conditions. Overall, the data indicate that the AI‐based method proposed herein could properly identify and score enzootic pneumonia‐like lesions without interfering with the slaughter chain routine. According to European legislation, the application of such a method avoids the handling of carcasses and organs, decreasing the risk of microbial contamination, and could provide further alternatives in the field of food hygiene.

Action Detection is a complex task that aims to detect and classify human actions in video clips. Typically, it has been addressed by processing fine-grained features extracted from a video classification backbone. Recently, thanks to the robustness of object and people detectors, a deeper focus has been added on relationship modelling. Following this line, we propose a graph-based framework to learn high-level interactions between people and objects, in both space and time. In our formulation, spatio-temporal relationships are learned through self-attention on a multi-layer graph structure which can connect entities from consecutive clips, thus considering long-range spatial and temporal dependencies. The proposed module is backbone independent by design and does not require end-to-end training. Extensive experiments are conducted on the AVA dataset, where our model demonstrates state-of-the-art results and consistent improvements over baselines built with different backbones. Code is publicly available at https://github.com/aimagelab/STAGE_action_detection.

In this paper we present a novel approach for bottom-up multi-person 3D human pose estimation from monocular RGB images. We propose to use high resolution volumetric heatmaps to model joint locations, devising a simple and effective compression method to drastically reduce the size of this representation. At the core of the proposed method lies our Volumetric Heatmap Autoencoder, a fully-convolutional network tasked with the compression of ground-truth heatmaps into a dense intermediate representation. A second model, the Code Predictor, is then trained to predict these codes, which can be decompressed at test time to re-obtain the original representation. Our experimental evaluation shows that our method performs favorably when compared to state of the art on both multi-person and single-person 3D human pose estimation datasets and, thanks to our novel compression strategy, can process full-HD images at the constant runtime of 8 fps regardless of the number of subjects in the scene.

Continual Learning has inspired a plethora of approaches and evaluation settings; however, the majority of them overlooks the properties of a practical scenario, where the data stream cannot be shaped as a sequence of tasks and offline training is not viable. We work towards General Continual Learning (GCL), where task boundaries blur and the domain and class distributions shift either gradually or suddenly. We address it through mixing rehearsal with knowledge distillation and regularization; our simple baseline, Dark Experience Replay, matches the network's logits sampled throughout the optimization trajectory, thus promoting consistency with its past. By conducting an extensive analysis on both standard benchmarks and a novel GCL evaluation setting (MNIST-360), we show that such a seemingly simple baseline outperforms consolidated approaches and leverages limited resources. We further explore the generalization capabilities of our objective, showing its regularization being beneficial beyond mere performance.

Nowadays, robots are heavily used in factories for different tasks, most of them including grasping and manipulation of generic objects in unstructured scenarios. In order to better mimic a human operator involved in a grasping action, where he/she needs to identify the object and detect an optimal grasp by means of visual information, a widely adopted sensing solution is Artificial Vision. Nonetheless, state-of-art applications need long training and fine-tuning for manually build the object's model that is used at run-time during the normal operations, which reduce the overall operational throughput of the robotic system. To overcome such limits, the paper presents a framework based on Deep Convolutional Neural Networks (DCNN) to predict both single and multiple grasp poses for multiple objects all at once, using a single RGB image as input. Thanks to a novel loss function, our framework is trained in an end-to-end fashion and matches state-of-art accuracy with a substantially smaller architecture, which gives unprecedented real-time performances during experimental tests, and makes the application reliable for working on real robots. The system has been implemented using the ROS framework and tested on a Baxter collaborative robot.

Depth cameras allow to set up reliable solutions for people monitoring and behavior understanding, especially when unstable or poor illumination conditions make unusable common RGB sensors. Therefore, we propose a complete framework for the estimation of the head and shoulder pose based on depth images only. A head detection and localization module is also included, in order to develop a complete end-to-end system. The core element of the framework is a Convolutional Neural Network, called POSEidon+, that receives as input three types of images and provides the 3D angles of the pose as output. Moreover, a Face-from-Depth component based on a Deterministic Conditional GAN model is able to hallucinate a face from the corresponding depth image. We empirically demonstrate that this positively impacts the system performances. We test the proposed framework on two public datasets, namely Biwi Kinect Head Pose and ICT-3DHP, and on Pandora, a new challenging dataset mainly inspired by the automotive setup. Experimental results show that our method overcomes several recent state-of-art works based on both intensity and depth input data, running in real-time at more than 30 frames per second.