373 citations · 463 across the 17 of their papers we have counts for
21 papers
UGO: Unified Architecture for General Multi-Object Tracking by Segmentation
Jer Pelhan, Alan Lukezic, Matej Kristan
General multi-object tracking (GMOT) tracks all instances of a user-specified category from a single first-frame exemplar. Prior work relies on bounding boxes and surrogate trainin…
CoDi -- an exemplar-conditioned diffusion model for low-shot counting
Grega Šuštar, Jer Pelhan, Alan Lukežič +1
Low-shot object counting addresses estimating the number of previously unobserved objects in an image using only few or no annotated test-time exemplars. A considerable challenge f…
Generalized-Scale Object Counting with Gradual Query Aggregation
Jer Pelhan, Alan Lukezic, Matej Kristan
Few-shot detection-based counters estimate the number of instances in the image specified only by a few test-time exemplars. A common approach to localize objects across multiple s…
Distractor-Aware Memory-Based Visual Object Tracking
Jovana Videnovic, Matej Kristan, Alan Lukezic
Recent emergence of memory-based video segmentation methods such as SAM2 has led to models with excellent performance in segmentation tasks, achieving leading results on numerous b…
A Distractor-Aware Memory for Visual Object Tracking with SAM2
Jovana Videnovic, Alan Lukezic, Matej Kristan
Memory-based trackers are video object segmentation methods that form the target model by concatenating recently tracked frames into a memory buffer and localize the target by atte…
A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation
Jer Pelhan, Alan Lukežič, Vitjan Zavrtanik +1
Low-shot object counters estimate the number of objects in an image using few or no annotated exemplars. Objects are localized by matching them to prototypes, which are constructed…
DAVE -- A Detect-and-Verify Paradigm for Low-Shot Counting
Jer Pelhan, Alan Lukežič, Vitjan Zavrtanik +1
Low-shot counters estimate the number of objects corresponding to a selected category, based on only few or no exemplars annotated in the image. The current state-of-the-art estima…
A New Dataset and a Distractor-Aware Architecture for Transparent Object Tracking
Alan Lukezic, Ziga Trojer, Jiri Matas +1
Performance of modern trackers degrades substantially on transparent objects compared to opaque objects. This is largely due to two distinct reasons. Transparent objects are unique…
A Low-Shot Object Counting Network With Iterative Prototype Adaptation
Nikola Djukic, Alan Lukezic, Vitjan Zavrtanik +1
We consider low-shot counting of arbitrary semantic categories in the image using only few annotated exemplars (few-shot) or no exemplars (no-shot). The standard few-shot pipeline…
Trans2k: Unlocking the Power of Deep Models for Transparent Object Tracking
Alan Lukezic, Ziga Trojer, Jiri Matas +1
Visual object tracking has focused predominantly on opaque objects, while transparent object tracking received very little attention. Motivated by the uniqueness of transparent obj…
A Discriminative Single-Shot Segmentation Network for Visual Object Tracking
Alan Lukežič, Jiří Matas, Matej Kristan
Template-based discriminative trackers are currently the dominant tracking paradigm due to their robustness, but are restricted to bounding box tracking and a limited range of tran…
DAL -- A Deep Depth-aware Long-term Tracker
Yanlin Qian, Alan Lukežič, Matej Kristan +2
The best RGBD trackers provide high accuracy but are slow to run. On the other hand, the best RGB trackers are fast but clearly inferior on the RGBD datasets. In this work, we prop…
D3S -- A Discriminative Single Shot Segmentation Tracker
Alan Lukežič, Jiří Matas, Matej Kristan
Template-based discriminative trackers are currently the dominant tracking paradigm due to their robustness, but are restricted to bounding box tracking and a limited range of tran…
CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark
Alan Lukežič, Ugur Kart, Jani Käpylä +4
A long-term visual object tracking performance evaluation methodology and a benchmark are proposed. Performance measures are designed by following a long-term tracking definition t…
Performance Evaluation Methodology for Long-Term Visual Object Tracking
Alan Lukežič, Luka Čehovin Zajc, Tomáš Vojíř +2
A long-term visual object tracking performance evaluation methodology and a benchmark are proposed. Performance measures are designed by following a long-term tracking definition t…
Object Tracking by Reconstruction with View-Specific Discriminative Correlation Filters
Ugur Kart, Alan Lukezic, Matej Kristan +2
Standard RGB-D trackers treat the target as an inherently 2D structure, which makes modelling appearance changes related even to simple out-of-plane rotation highly challenging. We…
Now you see me: evaluating performance in long-term visual tracking
Alan Lukežič, Luka Čehovin Zajc, Tomáš Vojíř +2
We propose a new long-term tracking performance evaluation methodology and present a new challenging dataset of carefully selected sequences with many target disappearances. We per…
FuCoLoT -- A Fully-Correlational Long-Term Tracker
Alan Lukežič, Luka Čehovin Zajc, Tomáš Vojíř +2
We propose FuCoLoT -- a Fully Correlational Long-term Tracker. It exploits the novel DCF constrained filter learning method to design a detector that is able to re-detect the targe…
Beyond standard benchmarks: Parameterizing performance evaluation in visual object tracking
Luka Čehovin Zajc, Alan Lukežič, Aleš Leonardis +1
Object-to-camera motion produces a variety of apparent motion patterns that significantly affect performance of short-term visual trackers. Despite being crucial for designing robu…
Discriminative Correlation Filter with Channel and Spatial Reliability
Alan Lukežič, Tomáš Vojíř, Luka Čehovin +2
Short-term tracking is an open and challenging problem for which discriminative correlation filters (DCF) have shown excellent performance. We introduce the channel and spatial rel…
Deformable Parts Correlation Filters for Robust Visual Tracking
Alan Lukežič, Luka Čehovin, Matej Kristan
Deformable parts models show a great potential in tracking by principally addressing non-rigid object deformations and self occlusions, but according to recent benchmarks, they oft…