arXiv subjects

Samia Bouchafa

Publications and source records attributed to Samia Bouchafa.

6 recordsLinked to original sources

arXiv2026

SVD-Prune: Spectral Vision Token Pruning under Extreme Compression

Vision-Language Models (VLMs) and Vision-Language-Action (VLA) models process hundreds of visual tokens, making visual representations a major computational bottleneck. We argue that visual token pruning should be formulated not as identifying individually important tokens, but as preserving the dominant low-rank spectral structure of the visual representation. Based on this formulation, we introduce SVD-Prune, a training-free framework that leverages Singular Value Decomposition (SVD) and statistical leverage scores for globally informed token selection. SVD-Prune is architecture-agnostic and can be applied at different stages of the multimodal pipeline, including encoder-side, decoder-side, and hybrid configurations Extensive experiments across VLM and VLA benchmarks demonstrate strong robustness under extreme compression. SVD-Prune reduces the visual token. SVD-Prune cuts the visual token budget by up to 99.3%, while, at a 16-token budget, encoder-side pruning lowers inference FLOPs by 82.4% and achieves a 4.18x end-to-end speedup, including pruning overhead.

cs.CV↗

arXiv2026

Decoder-Agnostic Token Merging for Vision Transformers: A Systematic Study of G2TM

Vision Transformers (ViTs) have achieved state-of-the-art performance across a range of computer vision tasks, mainly thanks to the self-attention mechanism. However, its complexity, increasing quadratically with the number of tokens, remains the major obstacle to ViT efficiency and deployment at scale. Token merging reduces this cost by aggregating redundant tokens. Yet existing methods are typically evaluated within a single architecture, leaving open whether their effectiveness stems from the merging mechanism itself or from the specific decoder they are paired with. We extend Graph-Guided Token Merging (G2TM), a single module inserted early in a ViT-based network, beyond its original Segmenter setting. We evaluate G2TM across three semantic segmentation frameworks (Segmenter, SETR, EoMT) and three decoder families (Linear, Transformer-, convolution-based), as well as standard ViT image classification. Our results show that G2TM's behavior and accuracy-efficiency trade-off are consistent across every tested architecture for a given backbone size, indicating that its effectiveness is a property of the encoder rather than the decoder. G2TM also generalizes well to image classification, achieving an even smaller degradation in accuracy compared to semantic segmentation. We further find that G2TM's optimal hyperparameters, resulting in a consistent drop in GFLOPs of 22-47% and an increase in throughput by up to 74% for segmentation models on ADE20K dataset, depend primarily on the backbone's pre-training recipe and on the target dataset, rather than on the decoder choice.

cs.CV↗

arXiv2026

Beyond Attention Scores: SVD-Based Vision Token Pruning for Efficient Vision-Language Models

Vision-Language Models (VLMs) have revolutionized multi-modal learning by jointly processing visual and textual information. Yet, they face significant challenges due to the high computational and memory demands of processing long sequences of vision tokens. Many existing methods rely on local heuristics, such as attention scores or token norms. However, these criteria suffer from positional bias and information dispersion, limiting their ability to preserve essential content at high pruning ratios and leading to performance degradation on visually detailed images. To address these issues, we propose SVD-Prune, a training-free, plug-and-play token pruning method based on Singular Value Decomposition. It decomposes the vision token feature matrix and selects the top-k tokens using statistical leverage scores, ensuring only tokens contributing most to the dominant global variance are preserved. Experiments show that SVD-Prune consistently outperforms prior pruning methods under extreme vision token budgets, maintaining strong performance even with 32 and 16 vision tokens.

cs.CV↗

arXiv2024

DH-PTAM: A Deep Hybrid Stereo Events-Frames Parallel Tracking And Mapping System

This paper presents a robust approach for a visual parallel tracking and mapping (PTAM) system that excels in challenging environments. Our proposed method combines the strengths of heterogeneous multi-modal visual sensors, including stereo event-based and frame-based sensors, in a unified reference frame through a novel spatio-temporal synchronization of stereo visual frames and stereo event streams. We employ deep learning-based feature extraction and description for estimation to enhance robustness further. We also introduce an end-to-end parallel tracking and mapping optimization layer complemented by a simple loop-closure algorithm for efficient SLAM behavior. Through comprehensive experiments on both small-scale and large-scale real-world sequences of VECtor and TUM-VIE benchmarks, our proposed method (DH-PTAM) demonstrates superior performance in terms of robustness and accuracy in adverse conditions, especially in large-scale HDR scenarios. Our implementation's research-based Python API is publicly available on GitHub for further research and development: https://github.com/AbanobSoliman/DH-PTAM.

cs.CV↗

arXiv2022

IBISCape: A Simulated Benchmark for multi-modal SLAM Systems Evaluation in Large-scale Dynamic Environments

The development process of high-fidelity SLAM systems depends on their validation upon reliable datasets. Towards this goal, we propose IBISCape, a simulated benchmark that includes data synchronization and acquisition APIs for telemetry from heterogeneous sensors: stereo-RGB/DVS, Depth, IMU, and GPS, along with the ground truth scene segmentation and vehicle ego-motion. Our benchmark is built upon the CARLA simulator, whose back-end is the Unreal Engine rendering a high dynamic scenery simulating the real world. Moreover, we offer 34 multi-modal datasets suitable for autonomous vehicles navigation, including scenarios for scene understanding evaluation like accidents, along with a wide range of frame quality based on a dynamic weather simulation class integrated with our APIs. We also introduce the first calibration targets to CARLA maps to solve the unknown distortion parameters problem of CARLA simulated DVS and RGB cameras. Finally, using IBISCape sequences, we evaluate four ORB-SLAM3 systems (monocular RGB, stereo RGB, Stereo Visual Inertial (SVI), and RGB-D) performance and BASALT Visual-Inertial Odometry (VIO) system on various sequences collected in simulated large-scale dynamic environments. Keywords: benchmark, multi-modal, datasets, Odometry, Calibration, DVS, SLAM

eess.IV↗

arXiv2021

PCA Event-Based Optical Flow for Visual Odometry

With the advent of neuromorphic vision sensors such as event-based cameras, a paradigm shift is required for most computer vision algorithms. Among these algorithms, optical flow estimation is a prime candidate for this process considering that it is linked to a neuromorphic vision approach. Usage of optical flow is widespread in robotics applications due to its richness and accuracy. We present a Principal Component Analysis (PCA) approach to the problem of event-based optical flow estimation. In this approach, we examine different regularization methods which efficiently enhance the estimation of the optical flow. We show that the best variant of our proposed method, dedicated to the real-time context of visual odometry, is about two times faster compared to state-of-the-art implementations while significantly improves optical flow accuracy.

cs.CV↗