AVID: A Weakly Aligned RGB-Infrared Dataset and Perturbation Benchmark for Tiny Aerial Targets

Dharini Raghavan; Vishnu Rao; Sethu Selvi S

cs.CVOct 29, 2026

AVID: A Weakly Aligned RGB-Infrared Dataset and Perturbation Benchmark for Tiny Aerial Targets

Authors: Dharini Raghavan, Vishnu Rao, Sethu Selvi S

Organizations: Georgia Institute of Technology · ARTPARK, Indian Institute of Science · Ramaiah Institute of Technology

Abstract

Long-range detection of small drones is complicated by their visual similarity to birds and by the limited detail available from ground-based cameras. Combining visible and infrared observations offers complementary information, but their usefulness depends on target size, cross-modal alignment, and image quality. We present AVID, a benchmark for studying these factors using real RGB-infrared bird-drone surveillance footage originating from the Drone Detection Dataset and curated for the IEEE ICIP 2025 Video and Image Processing Cup. AVID comprises 32,040 corresponding frame pairs from 115 videos, with 35,952 existing bird and drone annotations at $320\times256$ resolution. Targets occupy an unusually narrow size range: 68.1% have a geometric-mean bounding-box side of 8--16 pixels, and 99.0% meet the COCO small-object criterion. To characterize the geometric difficulty of combining the modalities, we introduce an object-relative alignment ratio and an alignment burden curve. In a sample of 300 frame pairs, 32.3% have estimated global translations greater than 5 pixels; among the 264 pairs with matching annotations, 16.3% exceed one median target side. We further introduce AVID-C, which simulates seven families of image degradation at five severity levels independently in RGB, infrared, or both. A fixed-feature analysis examines how scale, degradation, and added cross-modal shifts affect fitted bird-drone class overlap. At 12--16 pixels, the Bhattacharyya error bounds are 0.148 for RGB, 0.198 for infrared, and 0.046 for their combined features. These values describe separation in the specified feature and density models, not detector accuracy. Together, AVID and AVID-C support scale-aware, alignment-aware, and sensor-specific studies of tiny aerial targets.

Figures & tables

Dataset	Main target/view	Modalities	Pairing	Main characteristic
KAIST [ 8 ]	pedestrian, ground	RGB + thermal	synchronized	multispectral pedestrian detection
LLVIP [ 9 ]	pedestrian, ground	visible + IR	paired	low-light pedestrian imagery
M3FD [ 10 ]	multi-class, ground	visible + IR	paired	image fusion and object detection
VTUAV [ 14 ]	generic target, aerial	visible + thermal	paired video	visible-thermal object tracking
Anti-UAV300 [ 16 ]	UAV, surveillance	RGB + IR	both streams released	long-range UAV surveillance
MMFW-UAV [ 17 ]	fixed-wing UAV, air-to-air	visible + thermal	one-to-one	multi-sensor and multi-view observations

Table 1: Comparison of AVID with related visible-infrared and UAV datasets. The pairing column describes the correspondence provided between modalities. Paired or synchronized observations do not necessarily imply pixel-level alignment.

Split	RGB-IR pairs	Videos	Bird instances	Drone instances	Total instances
Train	28,790	102	14,113	18,335	32,448
Validation	3,250	13	1,358	2,146	3,504
Total	32,040	115	15,471	20,481	35,952

Table 2: Composition of AVID . Train and Validation contain disjoint source videos. Instance counts refer to the existing bird and drone bounding-box annotations.

Figure 1: Representative RGB-infrared frame pairs from AVID . The top row shows a drone, and the bottom row shows a bird. Red boxes mark the annotated targets. RGB and infrared provide corresponding observations but differ in target appearance and can retain spatial displacement.

Figure 2: Target size and RGB-infrared alignment in AVID . a , Distribution of geometric-mean target side lengths across the Train and Validation data. The highlighted 8–16 pixel range contains 68.1% of the annotated instances. b , Estimated RGB-infrared translation magnitude for 300 sampled frame pairs. The search is limited to $\pm 12$ pixels along each axis, corresponding to a maximum measurable translation magnitude of 16.97 pixels. c , Alignment burden for the 264 pairs with matching target annotations. The fractions exceeding one quarter, one half, and one full median target side are 27.3%, 26.1%, and 16.3%, respectively. The unit-scale alignment burden is $\widehat{\mathcal{B}}_{1}=0.265$ .

Family	Parameter	Severity levels 1–5
Gaussian noise	standard deviation	5, 10, 18, 28, 40
Speckle noise	multiplicative $\sigma$	0.05, 0.10, 0.18, 0.28, 0.40
Salt-and-pepper	corrupted pixel fraction	0.005, 0.01, 0.02, 0.04, 0.08
Gaussian blur	Gaussian $\sigma$ (px)	0.5, 1.0, 2.0, 3.5, 5.5
Uneven illumination	field strength	0.10, 0.20, 0.35, 0.50, 0.70
Motion blur	kernel length (px)	3, 7, 11, 17, 25

Table 3: AVID-C perturbation families and severity levels. The table lists the parameter values used for severity levels 1–5. Noise standard deviations are expressed in 8-bit intensity units. Camera instability includes image translation, rotation, and a small variation in scale.

Figure 3: Overview of the fixed-representation analysis used to study AVID . a , Annotated RGB and infrared target patches are converted to $24\times 24$ grayscale images and projected into PCA spaces learned from the clean data. b , Bird and drone feature distributions are compared using Bhattacharyya and Chernoff error bounds, together with a kernel-density estimate as a non-Gaussian comparison. c , Cross-modal displacement is measured relative to target size using $\rho=d_{\mathrm{align}}/s_{\mathrm{target}}$ . d , The resulting measurements are used to compare RGB, infrared, and combined features and to examine the effects of image degradation and increasing sensor misalignment.

Figure 4: Effect of target size on bird-drone class separation. a , Fitted Bhattacharyya error bounds across target-size intervals for RGB, infrared, and the combined representation using eight PCA components from each modality. The 0–8 pixel interval contains only 103 annotated instances and should be interpreted with caution. b , Fusion gap calculated using Eq. 11 for different target sizes. c , Comparison of Bhattacharyya bounds, optimized Chernoff bounds, and KDE estimates for targets between 12 and 16 pixels. d , Effect of the number of PCA components on the fitted error bounds. The combined representation uses $k$ components from each modality, giving a total of $2k$ features.

Figure 5: Effect of simulated image degradation on RGB and infrared class separation. Panels a–g show the fitted Bhattacharyya error bounds from clean images (severity 0) through severity level 5 for each of the seven perturbation families. h , Mean increase in the fitted error bound at severity level 5 relative to the corresponding clean baseline. i , Estimated noise levels in clean RGB and infrared images compared with Gaussian-noise severity level 1. The added noise has a standard deviation approximately 16.6 times the median RGB estimate and 7.8 times the median infrared estimate.

Shift budget $a$ (px)	Paired bound	Fusion gap	Gap retention
0	0.0370	0.1107	1.000
1	0.0383	0.1094	0.988
2	0.0414	0.1063	0.960
5	0.0472	0.1005	0.907
8	0.0507	0.0970	0.876
12	0.0542	0.0935	0.844

Table 4: Effect of additional infrared shift on fitted class separation. The shift budget

a

gives the maximum allowed displacement along each image axis, rather than a fixed radial displacement. Each nonzero condition is repeated over five trials.

Item	What to report
Input	RGB, infrared, or both modalities; original or resized image resolution
Data subset	Dataset split, video identifiers when relevant, and number of frames
Target scale	Geometric-mean bounding-box size intervals and number of instances
Corruption	Perturbation type, severity level, and affected modality or modalities
Alignment	Estimation method, search range, and object-relative alignment thresholds or burden curve
Aggregation	Results across target sizes and perturbation types, alongside any overall average

Table 5: Recommended reporting details for experiments using AVID and AVID-C . Reporting the evaluated data, target sizes, sensor conditions, and experimental settings helps make results easier to reproduce and compare.

Appendix figures & tables10 assets

Supplementary material from the paper’s appendix.

Appendix

Side (px)	0–4	4–8	8–12	12–16	16–20	20–24	24–28	28–32	32–48
Count	0	103	6,783	17,706	7,269	2,155	1,189	376	371

Appendix

Table 6: Distribution of target sizes in AVID . Target size is measured using the geometric-mean side length of each annotated bounding box.

Quantity	Value
Number of measured shifts	300
Pairs with matched annotations	264
Pairs without matched annotations	36
Mean estimated shift	4.66 px
Median estimated shift	1.00 px
Standard deviation	6.44 px

Appendix

Table 7: Summary of RGB-infrared alignment measurements. Global translation is estimated for 300 frame pairs, of which 264 have matching target annotations for the object-relative analysis. The maximum representable shift is determined by the translation search range and does not indicate the largest possible physical displacement between the cameras.

	0–8	8–12	12–16	16–20	20–24	24–28	28+
RGB	0.098	0.089	0.148	0.137	0.053	0.019	-
Infrared	0.040	0.143	0.198	0.217	0.133	0.039	-
RGB+IR (8+8)	$<$ 0.001	0.018	0.046	0.050	0.009	$<$ 0.001	-

Appendix

Table 8: Bhattacharyya error upper bounds across target sizes. The paired representation combines eight PCA components from RGB and eight from infrared. A dash indicates that the size interval does not meet the minimum sample requirement for both classes.

Representation	Bhattacharyya ( $s=0.5$ )	Optimized Chernoff	Optimal $s$	KDE
RGB	0.1484	0.1314	0.33	0.1650
Infrared	0.1979	0.1944	0.43	0.1774
RGB+IR	0.0457	0.0396	0.37	0.0472

Appendix

Table 9: Comparison of class-overlap estimates for targets between 12 and 16 pixels. Results are reported for the fitted Bhattacharyya bound, optimized Chernoff bound, and KDE-based estimate.

Representation	Total dimension	Bound	Composition
RGB PCA	8	0.1484	RGB only
Infrared PCA	8	0.1979	infrared only
RGB+IR	8	0.1396	4 RGB + 4 IR
RGB duplicated	16	0.1272	repeated RGB feature
Infrared duplicated	16	0.1913	repeated infrared feature
RGB+IR	16	0.0457	8 RGB + 8 IR

Appendix

Table 10: Effect of feature dimension on fitted class separation at 12–16 pixels. The equal-dimension and duplicated-feature comparisons help identify how feature dimension influences the estimated bounds.

Perturbation	RGB	Infrared
Gaussian noise	0.3292	0.1982
Speckle noise	0.3752	0.2215
Salt-and-pepper	0.3211	0.1889
Gaussian blur	0.1855	0.2041
Uneven illumination	0.3040	0.2446
Motion blur	0.2221	0.2155

Appendix

Table 11: Bhattacharyya error upper bounds at severity level 5 of AVID-C . The corresponding clean-image bounds are 0.1477 for RGB and 0.1623 for infrared.

Shift budget	Joint upper bound	Fusion gap	Gap retention
0	0.0370	0.1107	1.000
1	0.0383	0.1094	0.988
2	0.0414	0.1063	0.960
5	0.0472	0.1005	0.907
8	0.0507	0.0970	0.876
12	0.0542	0.0935	0.844

Appendix

Table 12: Effect of controlled infrared displacement on the paired feature representation. The shift budget specifies the maximum allowed displacement along each image axis rather than a fixed radial displacement. Nonzero conditions are evaluated over five trials.

Figure 6: Examples of simulated image degradation in RGB frames. Each row shows one of the seven perturbation families in AVID-C , with severity increasing from left to right. The corresponding parameter values are listed in Table 3 .

Figure 7: Examples of simulated image degradation in infrared frames. Each row shows one of the seven perturbation families across increasing severity levels, applied to the infrared imagery. The perturbation settings correspond to those used for the RGB examples.

Figure 8: Additional RGB-infrared examples from AVID . The montage includes paired frames from different source videos, showing variation in background, target appearance, target count, and thermal contrast. Red bounding boxes indicate the annotated targets.

Aerial Object Detection

Aerial object detection locates and classifies objects in drone and satellite imagery, often under scale, orientation, clutter, and compute constraints.

See all 126 papers

LER-YOLO: Reliability-Aware Expert Routing for Misaligned RGB-Infrared UAV Detection
May 20, 2026Liming Hou, Yueping Peng, Hexiang Hao +6Infrared Object Detection Multispectral Object Detection
Engineering University of PAP, Xi’an 710086, China · Unit Command Department, Officers College of PAP, Chengdu 610213, China
Bridging Severe Cross-Modal Misalignment: End-to-End Visible-Infrared Object Detection via Explicit Feature-Domain Affine Registration
Aug 11, 2026Qi Ming, Yuyang Wang, Mingjing Zhao +7Multispectral Object Detection Cross-Modal Alignment
Beijing University of Technology, China · Central South University, China · Beijing Electronics Science & Technology Institute +4
RDANet: Relative Degradation Aware Network for Infrared Small Target Detection
Aug 21, 2026Rui Liu, Jing Nie, Ying FuInfrared Small Target Detection Remote Sensing Small Object Detection
School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China · School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China

May 20, 2026J/K move · Enter open · S save

LER-YOLO: Reliability-Aware Expert Routing for Misaligned RGB-Infrared UAV Detection

Liming Hou, Yueping Peng, Hexiang Hao, Ji Wang+5

Engineering University of PAP, Xi’an 710086, China · Unit Command Department, Officers College of PAP, Chengdu 610213, China

Infrared Object Detection Multispectral Object Detection Cross-Modal Alignment Mixture of Experts Multimodal Sensor Fusion UAV Remote Sensing+1

Open paper AskPDF arXiv

Detecting small unmanned aerial vehicles from RGB-infrared remote-sensing pairs remains challenging due to tiny target scale, cluttered backgrounds, and spatial misalignment between heterogeneous sensors. Existing bimodal detectors often align or fuse features without assessing the reliability of local cross-sensor correspondence, allowing mismatch artifacts to propagate into the detection head. To address this issue, we propose LER-YOLO, a reliability-aware sparse mixture-of-experts framework for misaligned RGB-infrared UAV detection. LER-YOLO first introduces an Uncertainty-Aware Target Alignment module that resamples visible features toward the infrared reference and estimates a spatial reliability map. This reliability prior is then used by a Reliability-Guided Sparse MoE Fusion module to adaptively select k experts from RGB-dominant, infrared-dominant, and interactive fusion experts, enabling trustworthy cross-modal interaction while suppressing unreliable fusion. Experiments on the public MBU benchmark under a YOLOv5s-family protocol show that LER-YOLO achieves 89.7+/-0.2% AP50 over three independent seeds, with a best result of 89.9%. Extensive ablations, parameter-matched comparisons, synthetic-shift evaluations, and complexity analysis demonstrate that the gains mainly come from reliability-guided expert routing rather than increased model capacity.

Abstract

Aerial Object Detection

Explore similar work

LER-YOLO: Reliability-Aware Expert Routing for Misaligned RGB-Infrared UAV Detection

Bridging Severe Cross-Modal Misalignment: End-to-End Visible-Infrared Object Detection via Explicit Feature-Domain Affine Registration

RDANet: Relative Degradation Aware Network for Infrared Small Target Detection