Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Jorge P. BatistaSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works82 from public data
- 5,739
- Fully-Convolutional Siamese Networks for Object Tracking4,457
A basic tracking algorithm is equipped with a novel fully-convolutional Siamese network trained end-to-end on the ILSVRC15 dataset for object detection in video and achieves state-of-the-art performance in multiple benchmarks.
- Exploiting the Circulant Structure of Tracking-by-Detection with Kernels2,365
Using the well-established theory of Circulant matrices, this work provides a link to Fourier analysis that opens up the possibility of extremely fast learning and detection with the Fast Fourier Transform, which can be done in the dual space of kernel machines as fast as with linear classifiers.
- End-to-End Representation Learning for Correlation Filter Based Tracking1,469
This work is the first to overcome this limitation by interpreting the Correlation Filter learner, which has a closed-form solution, as a differentiable layer in a deep neural network, which enables learning deep features that are tightly coupled to the Cor correlation filter.
- Invariant Information Clustering for Unsupervised Image Classification and Segmentation1,005
A novel clustering objective is presented that learns a neural network classifier from scratch, given only unlabelled data samples, and discovers clusters that accurately match semantic classes, achieving state-of-the-art results in eight unsupervised clustering benchmarks spanning image classification and segmentation.
- The Visual Object Tracking VOT2016 Challenge Results776
The Visual Object Tracking challenge VOT2016 goes beyond its predecessors by introducing a new semi-automatic ground truth bounding box annotation methodology and extending the evaluation system with the no-reset experiment.
- The Visual Object Tracking VOT2014 Challenge Results588
The Visual Object Tracking challenge 2014, VOT2014, aims at comparing short-term single-object visual trackers that do not apply pre-learned models of object appearance and introduces a new dataset with full annotation of targets by rotated bounding boxes and per-frame attribute.
- Learning feed-forward one-shot learners500
This paper constructs the learner as a second deep network, called a learnet, which predicts the parameters of a pupil network from a single exemplar, and obtains an efficient feed-forward one-shot learner, trained end-to-end by minimizing a one- shot classification objective in a learning to learn formulation.
- 434
- Multi-modal Self-Supervision from Generalized Data Transformations197
The framework of Generalized Data Transformations is introduced to reduce several recent self-supervised learning objectives to a single formulation for ease of comparison, analysis, and extension, and allow a choice between being invariant or distinctive to data transformations, obtaining different supervisory signals, and derive the conditions that combinations of transformations must obey in order to lead to well-posed learning objectives.
- 186
- Long-Term Tracking in the Wild: A Benchmark184
The OxUvA dataset and benchmark for evaluating single-object tracking algorithms is introduced, offering the community a large and diverse benchmark to enable the design and evaluation of tracking methods ready to be used “in the wild”.
- 149
- Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image145
This work proposes Flash3D, a method for scene reconstruction and novel view synthesis from a single image which is both very generalisable and efficient, and achieves state-of-the-art results when trained and tested on RealEstate10k.
- 145
- Audio Retrieval With Natural Language Queries: A Benchmark Study137
This work employs three challenging new benchmarks to establish baselines for cross-modal text-audio and audio-text retrieval, and introduces the SoundDescs benchmark, which consists of paired audio and natural language descriptions for a diverse collection of sounds that are complementary to those found in AudioCaps and Clotho.
- 113
- Audio Retrieval with Natural Language Queries93
This work introduces challenging new benchmarks for text-based audio retrieval using text annotations sourced from the Audiocaps and Clotho datasets and employs these benchmarks to establish baselines for cross-modal audio retrieval, where the benefits of pre-training on diverse audio tasks are demonstrated.
- RbA: Segmenting Unknown Regions Rejected by All88
This paper proposes a novel outlier scoring function called RbA by defining the event of being an outlier as being rejected by all known classes, and shows that mask classification improves the performance of the existing outlier detection methods, and the best results are achieved with the proposed RbA.
- CASSPR: Cross Attention Single Scan Place Recognition83
CASSPR is proposed as a method to fuse point-based and voxel-based approaches using cross attention transformers using cross attention transformers, and surpasses the state-of-the-art by a large margin on several datasets.
- Text2Loc: 3D Point Cloud Localization from Natural Language81
This work introduces a novel neural network, Text2Loc, that fully interprets the semantic relationship between points and text and proposes a novel matching-free fine localization method, which completely removes the need for complicated text-instance matching and is lighter, faster, and more accurate than previous methods.
- 70
- Invariant Information Distillation for Unsupervised Image Segmentation and Clustering69
A simple loss based on information theory is used to extract meaningful representations directly from raw images, making it the first end-to-end unsupervised learning method that learns densely and efficiently (i.e. without sampling) for semantic segmentation.
- VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning66
Neuro-Symbolic Predicates is presented, a first-order abstraction language that combines the strengths of symbolic and neural knowledge representations and an online algorithm for inventing such predicates and learning abstract world models is outlined.
- Beyond the shortest path: Unsupervised domain adaptation by Sampling Subspaces along the Spline Flow58
- On Compositions of Transformations in Contrastive Self-Supervised Learning54
It is found that being invariant to certain transformations and distinctive to others is critical to learning effective video representations, improving the state-of-the-art for multiple benchmarks by a large margin, and even surpassing supervised pretraining.
- 53
- 29
- Discriminative Bayesian Active Shape Models28
This work presents a simple and very efficient solution to align facial parts in unseen images that improves on the state-of-the-art while using the same local detectors and shows that MOSSE filters further improve on these results.
- Semi-intrinsic Mean Shift on Riemannian Manifolds24
This work proposes a novel paradigm to intrinsically reformulate the mean shift on general Riemannian manifolds by embedding the Riem Mannian manifold into a Reproducing Kernel Hilbert Space by using a general and mathematically well-founded RiemANNian kernel function.
- Small Steps and Giant Leaps: Minimal Newton Solvers for Deep Learning23
A fast second-order method that can be used as a drop-in replacement for current deep learning solvers, and only requires two additional forward-mode automatic differentiation operations per iteration, which has a computational cost comparable to two standard forward passes and is easy to implement.
- 21
- 20
- Fast Training of Pose Detectors in the Fourier Domain20
The experiments show that training a sliding-window object detector and pose estimator can be sped up by orders of magnitude, for transformations as diverse as planar rotation, the walking motion of pedestrians, and out-of-plane rotations of cars.
- Towards Solving Fuzzy Tasks with Human Feedback: A Retrospective of the MineRL BASALT 2022 Competition18
The MineRL BASALT Competition on Fine-Tuning from Human Feedback at NeurIPS 2022 aimed to promote the development of algorithms that use human feedback as channels to learn the desired behavior.
- GST: Precise 3D Human Body from a Single Image with Gaussian Splatting Transformers17
This work combines 3D human pose and shape estimation with 3D Gaussian Splatting (3DGS), a representation of the scene composed of a mixture of Gaussians, to achieve near real-time inference of 3D human models from a single image without expensive diffusion models or 3D points supervision.
- Moving SLAM: Fully Unsupervised Deep Learning in Non-Rigid Scenes17
A new deep learning framework to decompose monocular videos into 3D geometry (camera pose and depth), moving objects, and their motions, with no supervision, with very competitive results on unsupervised odometry and depth prediction on KITTI.
- 17
- RapidVol: Rapid Reconstruction of 3D Ultrasound Volumes from Sensorless 2D Scans16
This work proposes RapidVol: a neural representation framework to speed up slice-to-volume US reconstruction, and demonstrates that further speed-up is achievable by reconstructing from a structural prior rather than from random initialisation.
- 15
- 13
- 11
- UniLoc: Towards Universal Place Recognition Using Any Single Modality10
This work develops a universal solution to place recognition, UniLoc, that works with any single query modality (natural language, image, or point cloud), and proposes a novel Self-Attention based Pooling module to evaluate the importance of instance descriptors when aggregated into a place-level descriptor.
- 10
- A Sound Approach: Using Large Language Models to Generate Audio Descriptions for Egocentric Text-Audio Retrieval9
This work considers the egocentric video setting and proposes three new text-audio retrieval benchmarks based on the EpicMIR and EgoMCQ tasks, and on the EpicSounds dataset, and confirms that LLMs can be used to determine the difficulty of identifying the action associated with a sound.
- UltraGauss: Ultrafast Gaussian Reconstruction of 3D Ultrasound Volumes8
This work introduces UltraGauss: the first ultrasound-specific Gaussian Splatting framework, extending view synthesis techniques to ultrasound wave propagation and derives an efficient rasterization boundary formulation for GPU parallelization and introduces a numerically stable covariance parametrization, improving computational efficiency and reconstruction accuracy.
- Dissecting Temporal Understanding in Text-to-Audio Retrieval8
This work dissects the temporal understanding capabilities of a state-of-the-art model for text-to-audio retrieval on the AudioCaps and Clotho datasets, and introduces a synthetic text-audio dataset that provides a controlled setting for evaluating temporal capabilities of recent models.
- A Light Touch Approach to Teaching Transformers Multi-view Geometry8
This work proposes a “light touch” approach, guiding visual Transformers to learn multiple-view geometry but allowing them to break free when needed, by using epipolar lines to guide the Transformer's cross-attention maps during training.
- Towards real-world navigation with deep differentiable planners8
This work trains embodied neural networks to plan and navigate unseen complex 3D environments, emphasising real-world deployment and is the first to successfully apply them to the difficult Active Vision Dataset, consisting of real images captured from a robot.
- Trafficloc: Localizing Traffic Surveillance Cameras in 3D Scenes6
The proposed TrafficLoc employs a novel Geometry-guided Attention Loss (GAL) to focus only on the corresponding geometric regions under different viewpoints during 2D-3D feature fusion and achieves new SOTA performance on KITTI and NuScenes datasets, demonstrating the superiority across both in-vehicle and traffic cameras.
- 6
- 6
- 5
- LangProp: A code optimization framework using Large Language Models applied to driving4
The first proof of concept of automated code optimization for autonomous driving in CARLA is demonstrated, showing that LangProp can generate interpretable and transparent driving policies that can be verified and improved in a metric-and data-driven way.
- 4
- HelloFresh: LLM Evaluations on Streams of Real-World Human Editorial Actions across X Community Notes and Wikipedia edits3
HelloFresh is introduced, based on continuous streams of real-world data generated by intrinsically motivated human labelers, and it is found that HelloFresh yields a temporally consistent ranking.
- Illusory Attacks: Information-Theoretic Detectability Matters in Adversarial Attacks3
It is concluded that agents must seek realism feedback to be robust to illusionary attacks, a novel form of attack that is consistent with the world model of the victim agent.
- 3
- Finding the Time to Think: Learning Planning Budgets in Real-Time RL2
This work introduces variable-delay real-time RL, where the agent chooses how long to deliberate at each decision point since the environment progresses, by training a lightweight gating policy on top of a planner to select state-dependent planning budgets.
- SCENES: Subpixel Correspondence Estimation With Epipolar Supervision2
This work removes the requirement of 3D structure from camera pose information, and replaces correspondence losses with epipolar losses, which encourage putative matches to lie on the associated epipolar line, and observes that this cue is sufficient for finetuning existing models on new data.
- Gradient Shape Model2
The Gradient Shape Model (GSM) is introduced, a novel approach that is able to align a similar CLM model without the need for any convolutions at all and allows a substantially reduction in the complexity/dimensionality of the data term, making it possible to compute a denser, more accurate, regression step per cascade level.
- 2
- CRAM: Large-Scale Video Continual Learning with Bootstrapped Compression1
This work expands current video CL benchmarks to large-scale settings, namely EpicKitchens-100 and Kinetics-700, storing thousands of relatively long videos in under 2 GB, and demonstrates empirically that the method outperforms prior art with a significantly reduced memory footprint.
- SOAP-RL: Sequential Option Advantage Propagation for Reinforcement Learning in POMDP Environments1
Evaluated against competing baselines, SOAP exhibited the most robust performance, correctly discovering options for POMDP corridor environments, as well as on standard benchmarks including Atari and MuJoCo, outperforming PPOEM, as well as LSTM and Option-Critic baselines.
- 1
- LoCUS: Learning Multiscale 3D-consistent Features from Posed Images1
It is found that it is possible to balance retrieval and reusability by constructing the retrieval set carefully, leaving out patches that map to far-away locations, and to regulate the scale of the learned features by adjusting the spatial tolerance for considering a retrieval to be positive.
- 1
- 1
- 1
- –
- PoseDreamer: Scalable and Photorealistic Human Data Generation Pipeline with Diffusion Models–
PoseDreamer is introduced, a novel pipeline that leverages diffusion models to generate large-scale synthetic datasets with 3D mesh annotations and results in better performance than combining real-world and synthetic datasets, demonstrating the complementary nature of the dataset.
- Rendering Novel Views of MRI Using 3D Gaussian Splatting–
Experiments show that across all stenosis conditions, resampled scans using Gaussian Splatting produce more accurate stenosis gradings compared to the raw scans which do not include the complete anatomy in-plane, as well as images resampled using Voxel Interpolation or Cubic B-spline resampling.
- Stale Diffusion: Hyper-realistic 5D Movie Generation Using Old-school Methods–
This work proposes Stale Diffusion, a method that solidifies and ossifies Stable Diffusion in a maximum-entropy state, and finds that naturally-accumulating errors have produced a novel entropy-maximising Stale Diffusion method, that can produce sleep-inducing hyper-realistic 5D video that is as good as one's imagination.
- –
- Interpretable Representation Learning from Videos using Nonlinear Priors–
A deep learning framework where one can specify nonlinear priors for videos that allow the model to learn interpretable latent variables and use these to generate videos of hypothetical scenarios not observed at training time is proposed.
- Large Language Models are Few-shot Publication Scoopers–
This work proposes a novel solution to the long-standing personal credit assignment problem and demonstrates the unparalleled potential of the algorithm to scoop groundbreaking findings with the insouciance of a seasoned researcher at a dessert buffet.
- –
- A 23 MW data centre is all you need–
It is discovered that individuals awake at this ungodly hour with access to a sufficiently powerful computer possess an opportunity for myriad forms of long-term linguistic 'lock in', and a tractable algorithm is proposed to ensure that, for the future of humanity, the British spelling of colour becomes the default spelling.
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.